List of Tables .................................................................................................................................. vi

List of Figures ...............................................................................................................................viii

Abbreviations.................................................................................................................................. ix

A note on transcriptions and glosses............................................................................................... xi

Foreword ........................................................................................................................................ xv

Author’s Preface ........................................................................................................................... xvi

Acknowledgements....................................................................................................................... xix

1 WORD, CLITIC, AND AFFIX ......................................................................................................... 19

2 WORDS AND WORD-FORMATION............................................................................................. 20

2.1 Verbs and verbal derivation ........................................................................................................... 20

2.1.1 Valency-neutral verbal affixes ................................................................................................. 31

2.1.1.1 Contained dxʷ(s)-............................................................................................................ 32

2.1.1.2 Inchoative -il ................................................................................................................... 36

2.1.1.3 Middle -b......................................................................................................................... 41

2.1.1.4 Autonomous action -agʷil............................................................................................... 46

2.1.1.5 Purposive -iluɬ................................................................................................................. 49

2.1.1.6 Method -áb...................................................................................................................... 51

2.1.1.7 Desiderative -ab .............................................................................................................. 52

2.1.1.8 Partitive ʔiɬ-..................................................................................................................... 54

2.1.2 Valency-increasing verbal affixes............................................................................................ 60

2.1.2.1 Internal causative -t ......................................................................................................... 62

2.1.2.2 External causative -txʷ .................................................................................................... 74

2.1.2.3 Diminished control causative -dxʷ.................................................................................. 87

2.1.2.4 Causative middle -b......................................................................................................... 94

2.1.2.5 Causative of activity -alikʷ ............................................................................................. 96

2.1.2.6 Allative applicative -c/-s ............................................................................................... 100

2.1.3 Secondary suffixes ................................................................................................................. 103

2.1.3.1 Dative applicative -yi-................................................................................................... 104

2.1.3.2 Middle applicative -bi-.................................................................................................. 109

2.1.3.3 Other secondary suffixes -di-, -i-.................................................................................. 119

2.1.4 Incorporative -əɬ..................................................................................................................... 124

2.1.5 Propriative bəs-...................................................................................................................... 127

2.1.6 Other verb-stem formatives.................................................................................................... 130

2.2 Nouns and nominal derivation ..................................................................................................... 131

2.2.1 Nominalizing prefix s-........................................................................................................... 132

2.2.2 Proclivitive dxʷs- ................................................................................................................... 138

2.2.3 Instrumental səxʷ-.................................................................................................................. 139

2.2.4 Proxy sixʷ- ............................................................................................................................. 142

2.2.5 Container xʷ-.......................................................................................................................... 143

2.2.6 Seasonal pəd- ......................................................................................................................... 145

2.2.7 Colour x&i- ............................................................................................................................... 145

2.2.8 Kinship prefixes ʔixʷ-, cixʷ-, and ciɬ-.................................................................................... 146

2.2.9 Implement -təd....................................................................................................................... 147

2.2.10 Relational -bid....................................................................................................................... 148

2.2.11 Other derivational affixes...................................................................................................... 150

2.3 Prepositions .................................................................................................................................... 151

2.3.1 Locative-temporal prepositions.............................................................................................. 153

2.3.1.1 Locative ʔal ‘at, on’ ...................................................................................................... 165

2.3.1.2 Centripetal dxʷʔal ‘to, towards, into’............................................................................ 174

2.3.1.3 Centrifugal tul’ʔal ‘from, out of’ .................................................................................. 182

2.3.1.4 Prolative liɬʔal ‘via, through’........................................................................................ 186

2.3.2 General preposition ʔə............................................................................................................ 187

2.4 Determiners .................................................................................................................................... 196

2.5 Adverbs and other predicate-modifiers....................................................................................... 223

2.5.1 Lexical adverbs and adverbial particles ................................................................................. 223

2.5.2 Locative adverbs .................................................................................................................... 230

2.5.3 Predicate particles .................................................................................................................. 238

2.5.4 Demonstrative adverbials....................................................................................................... 242

2.6 Interrogative/indefinite words...................................................................................................... 248

2.6.1 gʷat ‘who?’ ............................................................................................................................ 254

2.6.2 stab ‘what?’, stabaɬ ‘what kind of?’, and čədaɬ ‘which?’.................................................... 256

2.6.3 pədtab ‘when?’ ...................................................................................................................... 263

2.6.4 čad ‘where?’, dxʷčad ‘to where?’, tul’čad ‘from where?’, and liɬčad ‘which way?’ .......... 264

2.6.5 čal ‘how?’ .............................................................................................................................. 269

2.6.6 k’ʷid ‘how many? how much?’ .............................................................................................. 272

2.6.7 čayɬ ‘go for what reason?’ and ʔidigʷat ‘say what?’ ............................................................ 276

2.6.8 ʔəx&id ‘what happened?’.......................................................................................................... 278

2.7 Other word classes......................................................................................................................... 286

2.7.1 Personal and possessive pronouns ......................................................................................... 286

2.7.2 Quantifiers.............................................................................................................................. 292

2.7.3 Numerals ................................................................................................................................ 292

2.7.4 Directional particles ............................................................................................................... 302

2.7.5 Interjections and conjunctions................................................................................................ 309

2.8 Compounding................................................................................................................................. 310

3 PHRASE-LEVEL CLITICS............................................................................................................ 311

3.1 Tense and mood ............................................................................................................................. 311

3.1.1 Past tense tu=......................................................................................................................... 311

3.1.2 Habitual mood ƛ’u= ............................................................................................................... 311

3.1.3 Irrealis mood ɬu= ................................................................................................................... 311

3.1.4 Subjunctive mood gʷə=......................................................................................................... 311

iii

3.1.5 Negative mood lə= ................................................................................................................ 312

3.2 Progressive clitic lə= ..................................................................................................................... 312

3.3 Additive clitic bə=.......................................................................................................................... 312

4 LEXICAL SUFFIXES...................................................................................................................... 314

4.1 In verbs ........................................................................................................................................... 314

4.2 In nouns .......................................................................................................................................... 315

4.3 In numeral classification ............................................................................................................... 315

5 REDUPLICATION .......................................................................................................................... 319

5.1 Type I reduplication C

- ‘distributive’ ................................................................................. 320

5.2 Type II reduplication C

- ‘attenuative’ ................................................................................... 322

5.3 Type III reduplication -V

........................................................................................................ 328

5.3.1 Diminished effectiveness ....................................................................................................... 328

5.3.2 Intensivity............................................................................................................................... 329

5.3.3 Exclusivity.............................................................................................................................. 330

5.3.4 Plurality .................................................................................................................................. 331

5.4 Other plural reduplications .......................................................................................................... 331

5.5 Multiple reduplications ................................................................................................................. 333

6 VERBAL INFLECTION ................................................................................................................. 338

6.1 Aspect-marking.............................................................................................................................. 338

6.1.1 Simple aspects........................................................................................................................ 338

6.1.1.1 Imperfective aspect Ø- .................................................................................................. 340

6.1.1.2 Perfective aspect ʔu-...................................................................................................... 340

6.1.1.3 Stative aspect ʔas- ......................................................................................................... 340

6.1.2 Compound aspects — ləs- and ləcu- ..................................................................................... 353

6.2 Passive voice ................................................................................................................................... 353

7 REFERENTIAL EXPRESSIONS................................................................................................... 359

7.1 Deixis............................................................................................................................................... 359

7.2 Possessives ...................................................................................................................................... 359

7.3 Modification, attributives, apposition, and coordination........................................................... 361

7.3.1 Modification by adverbs ........................................................................................................ 362

7.3.2 Noun–noun attributive constructions ..................................................................................... 363

7.3.3 gʷəɬ ‘associative’ ................................................................................................................... 365

7.3.4 Appositive phrases ................................................................................................................. 374

7.3.5 Coordination........................................................................................................................... 377

7.4 Relative, headless relative, and nominalized clauses.................................................................. 383

7.4.1 Modifying and headless relative clauses................................................................................ 383

7.4.2 Nominalized clauses............................................................................................................... 389

7.4.2.1 s=nominals .................................................................................................................... 393

7.4.2.2 dəxʷ=nominals .............................................................................................................. 397

8 PREDICATIVE EXPRESSIONS.................................................................................................... 401

8.1 Grammatical relations................................................................................................................... 401

8.1.1 Subject-markers...................................................................................................................... 402

8.1.2 Object-markers....................................................................................................................... 414

8.1.3 Reflexive-markers .................................................................................................................. 418

8.1.4 Reciprocal-markers ................................................................................................................ 420

8.1.5 Subject, object, and direct complement noun phrases ........................................................... 421

8.1.6 Oblique objects....................................................................................................................... 424

8.1.7 Agentive complements........................................................................................................... 427

8.2 Verbal predicates ........................................................................................................................... 429

8.2.1 Monovalent intransitive clauses............................................................................................. 429

8.2.2 Transitive clauses ................................................................................................................... 429

8.2.3 Passive and bivalent intransitive clauses................................................................................ 430

8.2.4 Clauses with multiple objects................................................................................................. 430

8.2.5 Clauses with nominal predicate complements ....................................................................... 432

8.2.6 Clauses with verbal predicate complements .......................................................................... 440

8.2.7 Adverbs, locatives and adjunct phrases ................................................................................. 443

8.3 Non-verbal predicates ................................................................................................................... 443

8.3.1 Nominal and nominalized predicates ..................................................................................... 443

8.3.2 Other non-verbal predicates ................................................................................................... 446

8.4 Interrogatives ................................................................................................................................. 446

8.4.1 Polar questions ....................................................................................................................... 446

8.4.2 Information questions............................................................................................................. 450

8.5 Imperatives..................................................................................................................................... 458

8.6 Negation .......................................................................................................................................... 464

8.6.1 Adverbial negation................................................................................................................. 466

8.6.2 Existential negatives............................................................................................................... 470

8.6.3 Non-future negatives.............................................................................................................. 473

8.6.4 Future negatives ..................................................................................................................... 476

8.6.5 Negative imperatives.............................................................................................................. 479

8.7 Desideratives .................................................................................................................................. 482

8.8 Comparatives ................................................................................................................................. 487

8.9 Possession........................................................................................................................................ 491

9 COMPLEX SENTENCES ............................................................................................................... 495

9.1 Verb series ...................................................................................................................................... 495

9.2 Co-ordination ................................................................................................................................. 495

9.3 Subjunctive subordinate clauses and embedded interrogatives................................................ 495

9.4 Sentential complements................................................................................................................. 495

9.5 Adverbial clauses ........................................................................................................................... 495

10 LOCATIVE AND SPATIAL EXPRESSIONS ............................................................................ 499

10.1 Prepositional phrases..................................................................................................................... 499

10.2 Locative and directional adverbials............................................................................................. 499

10.3 Demonstrative adverbs.................................................................................................................. 499

11 COMMUNICATIVE STRUCTURE ............................................................................................ 500

11.1 Theme and Rheme ......................................................................................................................... 502

11.2 Focalization .................................................................................................................................... 511

11.2.1 Focalization with diɬ ‘just that one’ ...................................................................................... 511

11.2.2 Contrastive focalization using gʷəl....................................................................................... 521

11.3 Topic-alignment and reference-tracking..................................................................................... 525

11.3.1 Topic- and scene-setting constructions ................................................................................. 529

11.3.2 Topical object marker

-ag„id

............................................................................................... 539

Appendix 1: Forms based on free radicals................................................................................... 542

Appendix 2: Forms based on bound radicals............................................................................... 549

Appendix 3: Forms based on hypothetical radicals ..................................................................... 554

Appendix 4: Texts included in the Lushootseed Corpus ............................................................. 556

Bibliography ................................................................................................................................ 558

List of Tables

Table 1: Regular CVC(C) radicals................................................................................................ 20

Table 2: Regular CC radicals........................................................................................................ 21

Table 3: Radicals showing harmonic epenthesis .......................................................................... 23

Table 4: CəC radicals showing epenthesis.................................................................................... 23

Table 5: CC radicals showing epenthesis ..................................................................................... 23

Table 6: Radicals showing final voicing alternations................................................................... 24

Table 7: Patient-oriented radicals ................................................................................................. 26

Table 8: Agent-oriented radicals................................................................................................... 26

Table 9: Bivalent radicals ............................................................................................................. 28

Table 10: Stems formed with dxʷ(s)- ........................................................................................... 34

Table 11: Conventionalized lexical suffix combinations with dxʷ(s)-......................................... 35

Table 12: Lexical suffixes requiring dxʷ(s)- ................................................................................ 36

Table 13: Stems formed with -il ................................................................................................... 38

Table 14: Monovalent intransitive stems formed with -b............................................................. 42

Table 15: Middle forms with lexical suffixes............................................................................... 44

Table 16: Middle forms based on nouns....................................................................................... 44

Table 17: Stems formed with -agʷil ............................................................................................. 47

Table 18: Stems formed with -iluɬ................................................................................................ 50

Table 19: Valency-increasing affixes ........................................................................................... 61

Table 20: Stems requiring the -š allomorph of the internal causative .......................................... 66

Table 21: Internal causative stems with -š and -t forms ............................................................... 67

Table 22: Internal causative stems formed from free radicals...................................................... 68

Table 23: Internal causative stems formed from bound radicals .................................................. 70

Table 24: Inherently transitive stems containing the internal causative ....................................... 70

Table 25: Applicative uses of -t.................................................................................................... 72

Table 26: Verbs of taking and bringing formed with -txʷ ............................................................ 76

Table 27: Causative stems formed with -txʷ ................................................................................ 77

Table 28: Applicative uses of -txʷ ................................................................................................ 80

Table 29: Radicals with both -txʷ and -t forms ............................................................................ 84

Table 30: External causative forms requiring the stative aspect................................................... 85

Table 31: Stems formed with -dxʷ................................................................................................ 89

Table 32: Reflexive stems formed with -dxʷ................................................................................ 91

Table 33: Bivalent intransitive stems formed with -b .................................................................. 96

Table 34: Stems formed with -alikʷ ............................................................................................. 98

Table 35: Stems formed with -c.................................................................................................. 101

Table 36: Stems formed with -s.................................................................................................. 102

Table 37: Stems formed with -yi-d on monovalent bases .......................................................... 105

Table 38: Stems formed with -yi-d on bivalent bases ................................................................ 109

Table 39: Applicative uses of -bi-d ............................................................................................ 111

Table 40: Applicative stems formed with -bi-d expressing motive ........................................... 113

Table 41: Stems formed with -bi-d and lexical suffixes............................................................. 114

Table 42: Valency-neutral uses of -bi-d ..................................................................................... 115

vii

Table 43: Causativizing uses of -bi-d......................................................................................... 117

Table 44: Stems formed with -bi-d and the allative applicative................................................. 118

Table 45: Stems formed with -di-d............................................................................................. 119

Table 46: Stems formed with a lexical suffix and -i-d ............................................................... 122

Table 47: Nouns formed with s- from monovalent bases........................................................... 133

Table 48: Nouns formed with s- from bivalent intransitive bases.............................................. 134

Table 49: Nouns formed with s- from nominal bases................................................................. 135

Table 50: Unanalyzable nouns with s-........................................................................................ 136

Table 51: Nouns formed with dxʷs- ........................................................................................... 138

Table 52: Nouns formed with səxʷ- ........................................................................................... 140

Table 53: Nouns formed with xʷ- ............................................................................................... 144

Table 54: Nouns formed with pəd-............................................................................................. 145

Table 55: Basic colour terms formed with x&i- ............................................................................ 145

Table 56: Nouns formed with -təd.............................................................................................. 148

Table 57: Prepositions................................................................................................................. 151

Table 58: Lushootseed determiners ............................................................................................ 197

Table 59: Lexical adverbs........................................................................................................... 224

Table 60: Verbs derived from lexical adverbs............................................................................ 225

Table 61: Adverbial particles...................................................................................................... 228

Table 62: Locative adverbs......................................................................................................... 230

Table 63: Verb stems derived from locative adverbs ................................................................. 234

Table 64: Compound locative adverbs and related motion verbs............................................... 237

Table 65: Predicate particles....................................................................................................... 239

Table 66: Demonstrative adverbials ........................................................................................... 242

Table 67: Interrogative/indefinite words .................................................................................... 248

Table 68: Personal pronouns....................................................................................................... 287

Table 69: Possessive pronouns ................................................................................................... 289

Table 70: Cardinal numerals—plain series................................................................................. 292

Table 71: Cardinal numerals — human series............................................................................ 294

Table 72: Cardinal numerals—temporal-iterative series ............................................................ 294

Table 73: Compound numeral expressions for days and days of the week ................................ 296

Table 74: Compound words........................................................................................................ 310

Table 75: Simple aspects ............................................................................................................ 338

Table 76: Passive forms of internal causative stems ending in -š .............................................. 355

Table 77: Possessive markers ..................................................................................................... 359

Table 78: Types of modifying clauses........................................................................................ 383

Table 79: Matrix clause subject markers .................................................................................... 403

Table 80: Coordinate subject markers ........................................................................................ 406

Table 81: Subjunctive subject clitics .......................................................................................... 407

Table 82: Possessive subject-clitics ............................................................................................ 410

Table 83: Imperative subject markers......................................................................................... 413

Table 84: Summary table of subject-markers ............................................................................. 413

Table 85: Object-markers ........................................................................................................... 414

Table 86: Reflexive verbs without plain transitive forms........................................................... 420

viii

List of Figures

Figure 1: Affixal template of the Lushootseed verb stem............................................................. 30

Figure 2: Lushootseed spatial prepositions................................................................................. 154

Figure 3: Expressions for marine orientation.............................................................................. 237

Figure 4: Expressions for riverine orientation ............................................................................ 238

Figure 5: Ordering and possible combinations of reduplicands ................................................. 334

Abbreviations

√ = verbal radical

º = bound form

= = clitic boundary

– = affix boundary

• = lexical suffix boundary

˹ ˺ = phraseologized string

SMALL CAPS = semantic role

, = semantic actants

1, 2, 3 = first, second, third person

ACT = activity

ADD = additive

ADNM = adjunctive nominalizer

ALTV = allative applicative

ASSC = associative

ATTN = attenuative

AUTO = autonomous

CNN = connective

CNTNR = container

CNTRFG = centrifugal

CNTRPT = centripetal

COLL = collective

COORD = coordinative

CONT = continuous

COP = copula

CSMD = causative middle

CSTR = constructive

CTD = contained

DAT = dative applicative

DC = diminished control

DIM.EFF = diminished effectiveness

DIST = distal

DMA = demonstrative adverbial

DUB = dubitative

DSD = desiderative

DSTR = distributive

ECS = external causative

EXC = exclusivity

FEM = feminine

FOC = focus

HAB = habitual

HMN = human

ICS = internal causative

IMPF = imperfective

IMPL = implement

INCH = inchoative

INCRP = incorporative

INSTR = instrumental

INT = interrogative

INTJ = interjection

INTNS = intensivity

IRR = irrealis

MAP = middle applicative

MD = middle

MTHD = method

NEG = negative

NEGP = negative proclitic

NL = Northern Lushootseed

NM = nominalizer

NSPEC = non-specific

OBJ = object

PASS = passive

PFV= perfective

PL = plural

PO = possessive

PR = preposition

PRDCT = product

PRLV = prolative

PROC = proclititive

PROG = progressive

PROP = propriative

PROX = proximal

PRPV = purposive

PRTV = partitive

PTCL = particle

Pu = Puyallup

QTV = quotative

RCP = reciprocal

REFL = reflexive

REM = remote/hypothetical

RLNL = relational

SBJ = subjunctive

SBRD = subordinate

SCONJ = sentential conjunction

SEAS = seasonal

SG = singular

Sk = Skagit

SL = Southern Lushootseed

Sn = Snohomish

SPEC = specific

SS = secondary suffix

STAT = stative

Su = Suquamish

SUB = subject

TOBJ = topical object

TR = transitivizer

UNQ = unique

USk = Upper Skagit

A note on transcriptions and glosses

The orthography used for the presentation of data in this grammar follows in most respects

the orthography developed by Thom Hess, which is used in the majority of published materials

on Lushootseed and by the Lushootseed communities in their language programmes. This

orthography is based on a fairly standard Americanist version of the International Phonetic

Alphabet in which /ʃ/ is represented as /š/, /tʃ/ as /č/, /χ/ as /x%/, /dʒ/ as /ǰ/, /dz/ as /dᶻ/, /ts/ as /c/,

and /tɬ’/ as /ƛ’/. Phonemically long vowels are written as double letters (i.e., /a…/ = /aa/), and the

diacritic <ˑ> is used after vowels to indicate lengthening for rhetorical purposes. In some early

versions of the orthography (e.g., Hess 1967a; Hess & Hilbert 1976), <ɫ>) is used for the

voiceless lateral affricate rather than <ɬ>, and <j> is used rather than <dᶻ> for the voiced alveolar

affricate. The transcriptions used here will follow the current practices, and examples taken from

these earlier sources are amended to reflect the more modern orthography. The same practice has

been followed for sources written before the advent of the standardized orthography such as

Tweddell (1950); data drawn from these works are presented according to the standardized

spellings used in the Lushootseed Dictionary (Bates, Hess & Hilbert 1994).

Orthographic representations of Lushootseed words represent a standardized phonemic

spelling system reflecting pronunciation in careful speech rather than close phonetic

transcriptions of particular utterances; this spelling system ignores certain prosodically-driven

phonological alternations that would obscure the underlying morphological structure of words

and would lead to a variety of spellings of like items in different phonological contexts. Standard

Lushootseed orthography has also adopted two conventions which depart from the practice of

phonemic spelling. The first of these concerns the representation of feminine determiners

(Section 2.4) and demonstrative adverbials (2.5.4), which are pronounced with an initial

voiceless alveolar affricate (IPA /ts/) and written as <ts> rather than as <c> (the usual way of

xii

writing IPA /ts/ in Lushootseed). This practice, which is maintained in this grammar, reflects the

etymological presence of a morpheme *-s- used to mark feminine gender and underlines the

regularity of alternations such as ti ‘specific non-feminine’ ~ tsi ‘specific feminine’, kʷi ‘remote

non-feminine’ ~ kʷsi ‘remote feminine’. The second convention concerns the representation of

CV radicals such as √ʔa(h) ‘be there’ and √lu(h) ‘hear’, which are usually spelled in pedagogical

and popular materials with a final <h> (with or without parentheses). The final <h> was

introduced both to reflect the fact that the majority of Lushootseed verbal radicals are CVC in

shape, and also to make the etymology of stems formed on such radicals more apparent to

learners. However, because the use of this convention carries the risk of introducing into the

academic literature a potential inaccuracy in the phonemic representation of Lushootseed words,

the conventional <h> is not used in the examples presented in this grammar and, for the sake of

consistency, is removed from examples taken from published sources in which the convention is

followed.

The presentation of analyzed examples also differs from the practices followed in several

pedagogically-oriented publications such as the Lushootseed Readers (Hess 1995, 1998, 2006).

In these latter sources, several conventions are used to make the texts more accessible for

students who are following a written transcription while listening to a recording or who are

learning to parse Lushootseed words for the first time. Parentheses are used to “fill in” material

that is elided in predictable phonological contexts (e.g., /cutəb/ is given as “/cut(t)əb/” to alert

the reader to the loss of the suffix -t due to degemination). In this grammar, such material is not

included in text or in the first line of interlinearized examples, although it is reflected in the

second (parsed) line of interlinearized data (i.e., cutəb appears on line 1 of the interlinearization

and cut–t–əb on line 2). Pedagogical sources often use angle brackets (< >) to enclose

misspeakings, false starts, and stumbling on the part of speakers; this material is simply omitted

xiii

from the examples used here. Finally, materials prepared by Hess frequently appear with added

material enclosed in square brackets ([ ]) that was not part of the utterance as it was originally

recorded on tape. In most cases, these represent corrections made (by or in conjunction with a

native speaker) during transcription. When such editorial amendments clearly reflect the repair

of “performance” errors, they are included in the examples used here without mention; in a few

cases, the editorial amendments appear to be stylistic and to “correct” structures seen repeatedly

elsewhere throughout the corpus: in such instances the amendments are not included. In no case

is an amended example used for any purpose other than the illustration of a grammatical pattern

that is well (although perhaps less transparently) attested in other parts of the corpus, and such

examples are never used if they are the sole exemplar of a pattern or provide the key datum in

argumentation. All examples drawn from published sources are given with full citation by page

and line number, and readers interested in the scope and types of corrections made by

transcribers (or “errors” made by speakers) are urged to consult these sources, many of which

include audio recordings.

In a similar spirit, the full English glosses of example sentences follow the translations given

in published sources where this is appropriate and is not contraindicated by the necessities of

clear presentation. However, in numerous cases it was felt that the translation provided in the

source, while perhaps the most idiomatic or contextually appropriate gloss, was not the most

illuminating translation for the purposes of elucidating the underlying structure of the sentence or

for highlighting the morphosyntactic features under discussion. In a few cases, where the

Lushootseed construction strays too far beyond the bounds of idiomatic English, literal glosses

are provided in parallel to glosses in grammatical English. In other cases, the original gloss is

simply substituted for a gloss more suitable to the purposes of the discussion. As a full gloss is

an essentially heuristic device rather than an actual datum for the purposes of the

xiv

morphosyntactic analysis of data — the focus of the present work — it is hoped that the reader

will forgive this departure from normal citation practices. Those researchers who are interested

in the issues surrounding English-to-Lushootseed translation or who wish to explore the many

subtle semantic and cultural insights that can be gleaned by speakers’ choices of words or

phrasing are again urged to consult the original, contextualized sources.

Foreword

In the middle of the Nineteenth Century, George Gibbs, M.D., compiled a dictionary entitled

Niskwalli – English, English – Niskwalli. This is the earliest record of any kind available on

Dxʷləšucid (and it still merits careful study today).

Then, in the first half of the Twentieth Century several employees of the Smithsonian

Institution, various academics and a handful of interested lay people continued to document a

variety of lexical and grammatical facets of the language.

From June 1963 until August 1966 with funds made available to Prof. Laurence C.

Thompson by the National Science Foundation, I began my study of Dxʷləšucid. Prof

Thompson was my guide and mentor from then until 1983, when he suffered a major stroke.

More recently I have benefited from discussions with a variety of fellow Salishanists, in

particular Dr. Dawn Bates of Arizona State University and Dr. David Beck of the University of

Alberta.

Through all these years, my biggest contribution to Dxʷləšucid has been pedagogical. I

prepared a number of grammars for teaching the descendants of the Dxʷləšucid elders to read,

write and to speak the language (at least to some extent). Dr. Beck, however, with his supreme

mastery of a variety of linguistic theories here presents a grammar intended for linguists. It is for

academics, theoreticians, and graduate students majoring in linguistics. And it is a grammar I

would never have been able to write.

— Thom Hess Victoria,

14 August 2007

xvi

Author’s Preface

Writing a theoretically-informed descriptive grammar of a language is one of the most

profound challenges that a linguist can undertake. From the theoretical point of view, it is the

litmus test for the tools and premises that we bring to the task, and if either of these are unsuited

to the job, the language will break them, or the linguist will misrepresent the language. From the

practical point of view, the collection and untangling of the data, the careful recording and

probing of consultants, and the hours and years of wrestling with transcriptions, interpretations,

translations, and testable hypotheses represents a major investment of time, energy, career, and

life. So, from my perspective, when Thom Hess asked me to work on this grammar, it seemed to

me that I was being given the opportunity to take on a challenging and rewarding technical task,

while receiving essentially for free the fruits of his four decades of documentary work on the

language. This is an immense privilege, although it is also safe to say that without Thom’s

teaching, advice, and guidance, this grammar could never have been written (in spite of the

maddening modesty of his Foreward to this volume). In fact, in many ways Thom has written

much of this grammar. Drawing a line between what I know or “discovered” about Lushootseed

and what Thom, as a second-language speaker of Lushootseed and accomplished scholar, knows

and taught me and others is an impossible task that defies the conventions of scholarly citation. I

have done my best not to step too far over those bounds. If his name is not on the title page, it is

simply a reflection of the generosity and humility that has been the hallmark of his academic

career.

Writing a grammar that one hasn’t collected the primary data for, and one for which there are

no accessible native speakers to act as consultants, presents its own special challenges.

Fortunately, Lushootseed is one of the few disappearing languages of the Americas that has been

well and thoroughly documented through audio recordings, many (if not most) of which have

xvii

been carefully transcribed and put into a standardized orthography, largely through the efforts of

Thom Hess and Vi Taqʷš$blu Hilbert. The resulting corpus is a sizeable one for an indigenous

language that had existed exclusively in spoken form, apart from the sporadic efforts of

missionaries and linguists, until the last half of the 20

Century. The data for this grammar are

drawn primarily from a subset of this substantial documentary record, a corpus of 5,394 lines

(23,656 words) of the texts created by Hess and Hilbert — most recorded by Hess himself and

others recorded in the 1950s by Leon Metcalf (see Appendix 4) — analyzed and interlinearized

for the purposes of writing this grammar. Although small by the standards of corpora available

for languages like English (which run into the millions of words), this corpus provides a

sufficiently rich basis on which to make generalizations about what structures exist in the

language and how speakers use them — at least in the genre of traditional narrative, from which

all of the texts are drawn. The speakers all belong to a particular generation, the last generation

of Lushootseed-dominant native speakers, and as such represents a certain phase in the

language’s development. It is my feeling, confirmed by Thom Hess and others, that the speech of

this generation is a fair representation of Lushootseed as it was once spoken, although there are a

few constructions missing from their speech that are mentioned in the grammatical literature or

attested in the speech of the few elders of the previous generation who have been recorded.

The disadvantage of working with a corpus, as opposed to living speakers, is, of course, that

the corpus illustrates very well what people can do with the language, but does not reflect very

accurately what people can not do. The absence of a structure or a pattern from the corpus does

not necessarily demonstrate its unacceptability, and there are many places in the grammar where

questions have been left unanswered for lack of negative evidence. Nevertheless, in a corpus of

this size it is reasonable to expect that the major syntactic structures, lexical and morphological

patterns, and grammatical processes of the language will be fairly well represented, and the

xviii

absence of certain patterns from the most common context in the language can be taken as

evidence of, if not unacceptability, then strong dispreference. It is certainly possible that many of

the uncertainties created by gaps in the data will be resolved as the analyzed corpus grows, and

as historical-comparative evidence from the larger Salishan family is added; nevertheless, the

answers to many questions will likely never be found. Any grammar that can be written of a

language lost — even one as thoroughly and ably documented as Lushootseed — can be but a

pale echo of what it was when it was the living language of a thriving speech community.

xix

Acknowledgements

1 Word, clitic, and affix

Parts of speech (pred vs non-pred)

Clitics, bound clitics, and affixes

words: stressable, may head PhonP

S2 particles: unstressed, may go left or right depending on context

S2 clitics: always S2; always prosodically dependent; not restricted to hosts of a particular

lexical class; always en-

bound clitics: always either pro- or en-; always prosodically dependent and unstressed; not

restricted to hosts of a particular lexical class; may be iterated over an entire phrase

affixes: bound to stems of particular classes in fixed order; mark derivational meanings and

inflectional or quasi-inflectional categories of words

2 Words and word-formation

2.1 Verbs and verbal derivation

Probably one of the most notable features of Lushootseed (and other Salishan languages) is

the rich verbal morphology, particularly in the realm of derivation. The majority of Lushootseed

verb stems is built up out of fairly easily-analyzable elements based on monovalent radicals,

generally of the phonological shape CVC. A few of these are given in Table 1:

√ʔa ‘be there, exist’

√lil ‘far’

º√ʔəɬ ‘be eaten’

√p’il ‘be flat’

*bis ‘be selected’

º√qəd ‘fornicate’

º√c’əl ‘be defeated’

√q’əls ‘cook with steam’

º√c’əs ‘be pecked’

√q’ʷəl ‘be cooked, be ripe’

º√gəq ‘shining’

√q’ʷuʔ ‘be together with ’

º√gʷəč’ ‘be sought’

√šab ‘be dry’

√gʷəd ‘down’

√šidᶻ ‘launch sneak attack’

º√k’ʷalč’ ‘be bent backwards’

√t’agʷt ‘be on top’

º√kʷaʔ ‘be released’

º√x&əd ‘be pressed’

º√laʔ ‘be located’

º√xʷəb ‘be thrown’

º√ləc’ ‘have come down on’

*xʷak’ʷ ‘be tired’

√lək’ʷ ‘be eaten’

º√yəc ‘report’

Table 1: Regular CVC(C) radicals

A number of these radicals (√) are attested in independent form, appearing in sentences inflected

for aspect, person, and number, but without further derivational morphology. Others are bound

radicals (º) which are unattested in independent form but are productively used in the formation

of verb stems (e.g., º√c’əl ‘be defeated’ — c’əlalikʷ ‘win out over someone’, c’əld ‘defeat

someone’, c’əldxʷ ‘manage to defeat someone’).

A third group of radicals illustrated in Table 1

Note that this radical is often written with an orthographic final /h/ in texts and pedagogical grammars, both to

make it clear that it is a radical and to distinguish it from other [ʔa] sequences which are parts of other morphemes.

However, the /h/ is never actually pronounced and so it will not be included in cited forms used in this grammar,

including those taken from published texts in which the orthographic /h/ is used.

This radical is more properly grouped with the class of locative adverbs, though it is included here because, in

terms of its phonological and derivational properties, it patterns consistently with verbal radicals.

Note that it is possible that some of the forms marked as bound radicals here may in fact be potentially free

elements; radicals are marked as bound in this grammar if they are not attested as independent forms in the corpus

used for the present study or exemplified as independent forms in the Lushootseed Dictionary (Bates, Hess &

Hilbert 1994). Because of the pragmatically-odd meanings of some of the bare radicals, particularly the patient-

oriented radicals (see discussion below), it may be the case that the independent use of some of these forms is

possible but textually infrequent (cf. Gerdts 2006, who reports a great deal of success in the deliberate elicitation of

previously-unattested bare radicals in Halkomelem Salish).

are those marked *, which represent radicals that are both unattested as free forms and which do

not appear to be productively used as bases for derivation. This category includes forms that

appear only as a part of another productive derivational base (e.g., *xʷak’ʷ ‘be tired’, which

appears as part of a fossilized inchoative form √xʷak’ʷil ‘get tired’, itself a productive base for

forms such as xʷak’ʷilbid ‘become disaffected with someone’, and xʷak’ʷis ‘get tired of

someone’), and forms which appear in a single stem with an easily-analyzed affix whose

meaning is consistent with the meaning or syntactics of the derivational morpheme (e.g., *bis ‘be

selected’, the historical base of bisəd ‘select something’). In addition to radicals with the

canonical CVC(C) shape, there are a large number of CVCVC (e.g., √čəbaʔ ‘be loaded down

with something’, √həliʔ ‘be alive’, √p’ayəq ‘carve canoe’, √wiliq’ʷ ‘make an enquiry’) and a

few CVCV radicals (√bali ‘be forgetful’, √čubə ‘go inland’, √sula ‘be in the middle’); more

complex radicals are also attested (√ʔigʷəɬa ‘climb tree’, √hədʔiw’ ‘be inside a house’, √təlawil

‘run a distance’, √xʷiʔxʷiʔ ‘hunt for something, forage for something’). Many of this last group

appear to be fossilizations of diachronically analyzable strings formed through affixation or

reduplication.

A few radicals surface in independent form and in predictable prosodic contexts as CəC, but

do not have the schwa in the presence of potentially stress-bearing derivational affixes:

√šq ‘be high’

šəqəd ‘move  up high’

šəqlax&adəb ‘raise arms’

šqil ‘rise up’

º√ts ‘be punched’

təsəd ‘punch ’

tsalikʷ ‘hammer , pound ’

√t’q’ ‘be patched (with stickum)’

t’!q’əd ‘patch ’

t’q’abid ‘put stickum on ’

º√x&ƛ’ ‘be bitten’

x&əƛ’əd ‘bite ’

x&ƛ’alikʷ ‘bite into ’

º√x&q ‘be wrapped, be tied’

x&əqəd ‘wrap ’

Table 2: Regular CC radicals

This radical is more properly grouped with the class of locative adverbs, though it is included here because, in

terms of its phonological and derivational properties, it patterns consistently with verbal radicals.

The first of these radicals, √šq ‘high’, surfaces as [šəq] when stress is required on the stem (as in

šəˊqəd ‘move something up high’) and when it is required to break up lengthy consonant

clusters, but as [šq] when suffixation provides a full non-schwa vowel to carry stress. A similar

pattern is observed with the other forms in the table. There is some variation among speakers as

the to treatment of the schwa in these radicals, and in some cases the predicted presence/absence

of schwa is not found in all of the forms derived therefrom.

Regular CVC(C) radicals show no base-allomorphy when undergoing derivation; however,

there is a large group of CVC radicals which undergo a process of final harmonic vowel-

epenthesis in the presence of either the internal causative suffix -t (Section 2.1.2.1) or the

causative middle suffix -b (2.1.1.3). Some of these radicals, along with their -t or -b forms, are

given in Table 3:

√ʔil ‘sing’

ʔilid ‘sing ’

√ʔix&ʷ ‘be thrown; have thrown to’

ʔix&ʷid ‘throw  away’

√bap ‘be busy’

bapad ‘pester ’

√č’axʷ ‘be hit with a stick’

č’axʷad ‘club ’

√caq’ ‘be speared, be impaled’

caq’ad ‘spear ’

√cil ‘be dished up’

cilid ‘dish  out’

√dᶻakʷ ‘be shaky, be shaking’

dᶻakʷad ‘rock ’

√daʔ ‘be named’

daʔad ‘name ’

√dᶻix& ‘be broken down, be fallen apart’

dᶻix&id ‘break  down, take  apart’

√gʷi ‘make an invitation’

gʷiid ‘invite , call to ’

√huy ‘be done, be made, be finished’

huyud ‘make ’

√k’ʷil ‘peek’

k’ʷilid ‘peek at ’

√ƛ’iq ‘emerge’

ƛ’iqid ‘take  out from within’

√ɬaq’ ‘be fallen, be lying down’

ɬaq’ad ‘put  down’

√ɬič’ ‘get cut with knife’

ɬič’id ‘slice ’

√pus ‘be hit by  (missile)’

pusud ‘throw at ’

√q’axʷ ‘be frozen’

q’axʷad ‘freeze ’

√q’il ‘be aboard’

q’ilid ‘put  on board’

√q’ʷib ‘be disembarked, be unloaded’

q’ʷibid ‘unload  from conveyance’

√qʷat ‘be lying; snow falls’

qʷatad ‘lay  out’

qʷataš ‘lay  out’

√qʷib ‘be ready’

qʷibid ‘prepare ’

√šub ‘disappear’

šubud ‘make  disappear; massacre ’

√šul ‘be in, be under’

šulud ‘pass underneath ’

Like the radical √ʔa, the radical √gʷi and its derivatives are often written with an orthographic /h/ which is not

pronounced and which is not included in the cited form in this grammar.

This radical also refers to the return of anadromous fish.

The two transitive forms of this radical are based on different allomorphs of the internal causative suffix, [-t] and

[-š]. See Section 2.1.2.1 for further discussion.

º√tux&ʷ ‘be stretched’

tux&ʷud ‘stretch ’

√t’uc’ ‘be shot, fired on’

t’uc’ud ‘shoot  (target)’

√x&al ‘be written’

x&alad ‘write ’

√x&ʷaq’ʷ ‘be worried, be preoccupied’

x&ʷaq’ʷad ‘be concerned about ’

√yiq’ ‘be worked into tight place’

yiq’ib ‘make  (baskets)’

Table 3: Radicals showing harmonic epenthesis

Radicals belonging to this class epenthesize a harmonic copy of the stem vowel before the

derivational suffix. Although membership in this class of epenthesizing stems is not predictable,

all of them are CVC and none of them has the form C$C. Many C$C radicals also undergo final

epenthesis, as shown in Table 4:

√bəč ‘be lying, be fallen from standing’

bəčad ‘set  down’

bəčaš ‘set  down’

√dᶻəkʷ ‘travel, wander’

dᶻəkʷud ‘lead  astray, mislead ’

√gʷəx& ‘be untied’

gʷəx&ad ‘untie ’

√kʷəd ‘be held, be taken’

kʷədad ‘take ’

º√ləxʷ ‘be stabbed, be cut’

ləxʷud ‘stab ’

Table 4: C!C radicals showing epenthesis

These radicals epenthesize either /a/ or /u/, depending on their final consonant. A similar pattern

is seen in a small group of CC radicals:

º√pkʷ ‘be broken off’

pqʷud ‘break off ’

º√p’t’ ‘be stored’

p’t’ad ‘store , tidy ’

º√qʷc ‘slide, slip’

qʷəcad ‘slide ’

√q’p ‘form a lump; cramp up (muscle)’

q’pud ‘gather up’

º√q’p’ ‘be compensated’

q’p’ud ‘pay ’

*√q’x& ‘be insulted’

q’x&ad ‘insult ’

º√tq ‘be closed’

tqad ‘close , block  off’

º√tx&ʷ ‘be pulled’

tx&ʷud ‘pull on ’

º√xʷt’ ‘be fallen, be descended’

xʷt’ad ‘take  down’

Table 5: CC radicals showing epenthesis

As shown by the examples in Table 4 and Table 5, these radicals take an epenthetic /u/ if they

end in a bilabial or labialized consonant, and /a/ otherwise. The use of /a/ as an epenthetic vowel

with C$C and CC radicals is also seen in certain reduplicative environments (specifically, with a

sub-type of Type I reduplication, Section 5.1). Once again, final-vowel epenthesis with radicals

of this class takes place only in the presence of the internal causative and the causative middle

suffixes.

A few radicals show a voicing alternation in the final obstruent of the stem:

√c’aʔkʷ ‘be washed’

c’agʷad ‘wash ’

√čaʔkʷ ‘seaward’

čagʷəd ‘take  out to sea’

čagʷil ‘get out to sea’

čaʔkʷdxʷ

‘manage to get  to sea’

√čac ‘be hidden’

čadᶻil ‘become hidden, hide self’

čactxʷ ‘hide ’

√dəkʷ ‘be inside’

dəgʷad ‘put  inside’

º√dikʷ ‘be advised’

digʷid ‘advise ’

dikʷdxʷ

‘instruct ’

√hikʷ ‘big’

higʷəd ‘uphold , support ’

º√ƛ’akʷ ‘be stitched’

ƛ’agʷəb ‘make  (mat)’

ƛ’akʷtəd ‘cattail needle’

√takʷ ‘be bought’

tagʷš ‘buy ’

√təč ‘roll off, tumble down’

təǰəd ‘roll ’

º√t’ukʷ ‘be measured’

t’ugʷud ‘figure  out’

t’ukʷtəd ‘tape measure’

√x&ʷəc ‘be sharp’

dxʷx&ʷədᶻəb ‘be tart, be strong (coffee)’

Table 6: Radicals showing final voicing alternations

With these radicals, the final consonant is voiceless when in ultimate final position or when

followed by a derivational affix other than the inchoative -il (Section 2.1.1.2), the internal

causative -t, or the middle -b. Note that some members of this class also undergo final-epenthesis

(e.g., √c’aʔkʷ ‘be washed’ > c’agʷad ‘wash something’, √dəkʷ ‘be inside’ > dəgʷad ‘put

something inside’).

Two of these radicals, √c’aʔkʷ ‘be washed’ and √čaʔkʷ ‘seaward’, also lose

the glottal stop following the vowel; the same pattern is observed of the adverb √haʔkʷ ‘for a

long time’, which has the form hagʷəxʷ ‘finally, at last’, a lexicalized combination of the radical

with the temporal enclitic =əxʷ ‘now’ (Section *.*).

With only a few exceptions (see Table 9 below), Lushootseed radicals are monovalent and

intransitive, and require valency-increasing morphology (Section 2.1.1.7) to form verb stems

with a valency of greater than one. One consequence of this that has attracted a good deal of

attention in the literature (e.g., Hess 1995; Beck 1996, 2000b) is that Lushootseed has almost no

underived transitive verbs. What are transitive verbs in most languages are formed by derivation

This radical is more properly grouped with the class of adverbs, though it is included here because, in terms of its

derivational properties, it patterns consistently with verbal radicals.

The schwas in some of the derived forms in Table 6 (e.g., čagʷəd ‘take something out to sea’, ƛ’agʷəb ‘make

something (mat)’, təǰ əd ‘roll something’) belong to allomorphs of the internal causative or middle suffixes. See

Sections 2.1.2.1 and 2.1.2.4 below.

from a very large set of monovalent radicals which Hess (1995) describes as being “patient-

oriented” in the sense that their syntactic subject expresses the semantic PATIENT or ENDPOINT of

a semantically-transitive event rather than the AGENT. Consider the example in (1):

(1) a. ʔuɬič’ čəd

ʔu–ɬič’ čəd

PFV–be.cut.with.knife 1SG.SUB

‘I got cut with a knife’

b. ʔuɬič’id čəd tə sqʷiqʷali

ʔu–ɬič’i–d čəd tə sqʷiqʷali

PFV–be.cut.with.knife–ICS 1SG.SUB NSPEC hay

‘I started to cut hay (with a blade)’

(Bates, Hess & Hilbert 1994: 146)

In (1a), the verbal radical √ɬič’ ‘be cut with a knife’ —in spite of expressing an event high on

Hopper & Thompson’s (1980) scale of semantic transitivity — can take only a single syntactic

argument, a subject expressing the PATIENT of the event. In order to express an AGENT, it is

necessary to apply a valency-increasing suffix such as the internal causative -t, shown in (1b).

Thus, a patient-oriented radical in itself is focused on the final state achieved or the change

undergone by a PATIENT or ENDPOINT of an event, rather than on the cause of that state or the

involvement of an AGENT. A number of patient-oriented radicals are given in Table 5:

º√ʔadᶻq ‘be met’

º√gʷəlal ‘be hurt’

º√suxʷt ‘be recognized’

º√ʔay’ ‘be traded’

√gʷəx& ‘be untied’

√takʷ ‘be bought’

º√ʔəɬ ‘be eaten’

º√hay ‘be known’

º√təq’ ‘be slapped’

º√ʔuq’ʷ ‘be unplugged’

º√hiq ‘be pushed’

º√ts ‘be punched’

º√biƛ’ ‘be smashed, be crumbled’

º√k’aw ‘be chewed’

º√tudəq ‘be enslaved’

√caq’ ‘be speared, be impaled’

º√k’aw ‘be bumped’

º√tul ‘be interpreted’

√cil ‘be dished up’

º√kʷaxʷ ‘be helped’

º√tup ‘be pounded’

√c’aʔkʷ ‘be washed’

º√kʷaʔ ‘be released’

º√tux&ʷ ‘be stretched’

º√c’əl ‘be defeated’

√kʷəd ‘be held, be taken’

º√tx&ʷ ‘be pulled’

º√c’əs ‘be pecked’

√lək’ʷ ‘be eaten’

√t’q’ ‘be patched (with stickum)’

º√č’aʔ ‘be dug up’

º√ləxʷ ‘be stabbed, be cut’

√t’uc’ ‘be shot, fired on’

√č’axʷ ‘be hit with a stick’

º√t’ukʷ ‘be measured’

º√č’ədᶻ ‘be stalked’

º√ɬal ‘be removed from fire’

º√xʷac ‘be hoisted’

√daʔ ‘be named’

√ɬič’ ‘be cut with knife’

º√xʷəb ‘be thrown’

º√c’ix& ‘be fried’

√ɬid ‘be tied’

º√xʷš ‘be thrown’

º√c’uqʷ ‘be sucked on’

º√ɬuq’ʷ ‘be peeled’

√x&al ‘be written’

√čac ‘be hidden’

º√ƛ’akʷ ‘be stitched’

º√x&əd ‘be pressed’

√čal

‘be overtaken’

º√ƛ’ip’ ‘be compressed’

º√x&ib ‘be grabbed, clawed’

º√dikʷ ‘be advised’

º√p’ic’ ‘be wrung out’

º√x&q ‘be wrapped, be tied’

º√dᶻub ‘be kicked’

º√puʔ ‘be blown on’

º√x&ʷadᶻ ‘be injured’

√gəlk’ ‘be wound, be tangled’

√qiq’ ‘be confined’

º√gʷəč’ ‘be sought’

√qʷal ‘be marked, be painted’

√yiq’ ‘be worked into tight place’

Table 7: Patient-oriented radicals

With only one or two exceptions, patient-oriented radicals of this semantic type form their

transitive counterparts with the internal causative -t (Section 2.1.2.1); many of them also take the

diminished control suffix -dxʷ (2.1.2.3) and the causative of activity -alikʷ (2.1.2.4). The same is

true of a number of unaccusative radicals whose sole actant is non-agentive, but not entirely or

necessarily patient-like. These include radicals corresponding to what are labile verbs in English

(e.g., √hud ‘burn’, √k’ʷəɬ ‘pour out, spill out’, √q’axʷ ‘freeze, be frozen’), verbs of location

(√bəlx&ʷ ‘be beyond’, √č’it ‘nearby’, √dəkʷ ‘be inside’), and some states (º√ǰuʔ ‘be glad’, √qəɬ

‘be awake’, √x&ʷaq’ʷ ‘be worried, be preoccupied’) and processes (√ɬač’ ‘go out (fire)’, √ƛ’iq

‘emerge’).

A slightly smaller group of radicals falls into the category of AGENT-oriented verbs whose

subjects express semantic AGENTS or agent-like event-participants:

√ʔəƛ’ ‘come’

º√gʷuh ‘bark (dog)’

√qʷiʔad ‘yell’

√ʔibəš ‘travel, walk’

√kiis ‘stand up’

√qʷuʔqʷa ‘have a drink’

√ʔigʷəɬa ‘climb tree’

√kʷatač ‘climb’

√saq’ʷ ‘fly’

√ʔil ‘sing’

º√k’ʷəƛ’ ‘miss’

√šub ‘disappear’

√ʔux&ʷ ‘go’

√k’ʷit’ ‘go down to shore’

º√tatab ‘speak’

√cut ‘speak’

√lab ‘appear’

√tay ‘go raiding’

º√c’əb ‘clear land’

√lax& ‘recall, remember’

√təč ‘roll off, tumble down’

º√c’ic’əyikʷ ‘wink’

√ɬaʔ ‘arrive at a specific place’

º√t’iwiɬ ‘practice religion’

√č’əlp ‘turn’

√ləq ‘listen’ (Sk)

√t’uk’ʷ ‘go home’

√dᶻal ‘turn around, turn over’

√p’ayəq ‘carve canoe’

√wiliq’ʷ ‘make an enquiry’

√gʷah ‘accompany, go along’

√p’əq’ ‘drift’

º√yəhub ‘tell legend’

√gʷax&ʷ ‘take a stroll’

º√qəd ‘fornicate’

√yəy’duʔ ‘swing in a swing’

√gʷi ‘make an invitation’

√q’əlb ‘camp out’

Table 8: Agent-oriented radicals

The majority of these verbs are verbs of motion (e.g., √ʔəƛ’ ‘come’, √kʷatač ‘climb’, √yəy’duʔ

‘swing in a swing’) or activity (º√c’əb ‘clear land’, √p’ayəq ‘carve canoe’, √tay ‘go raiding’,

º√yəhub ‘tell legend’). Verbs of the latter type tend to express culturally-important activities and

cannot take an object without further derivation. Like the patient-oriented radicals, agent-

oriented radicals take a wide range of valency-increasing affixes to form transitive and bivalent

intransitive verbs, although as a set they are less consistent in their derivational possibilities. A

number of these form transitive stems with the internal causative (e.g., √ʔil ‘sing’ > ʔilid ‘sing

something’, √dᶻal ‘turn around, turn over’ > dᶻalqəd ‘turn something around’, √wiliq’ʷ ‘make an

enquiry’ > wiliq’ʷid ‘ask someone’); however, more of them form transitive verbs with the

external causative -txʷ (Section 2.1.2.2). This is especially true of the verbs of translational

motion (e.g., √ʔəƛ’ ‘come’ > ʔəƛ’txʷ ‘bring something’, √ʔux&ʷ ‘go’ > ʔux&ʷtxʷ ‘take something’,

√t’uk’ʷ ‘go home’ > t’uk’ʷtxʷ ‘take something home’), though a number of radicals from other

semantic classes also appear with this suffix (√kiis ‘stand up’ > kiistxʷ ‘stand something up’,

√lax& ‘recall, remember’ > lax&txʷ ‘remind someone’, º√yəhub ‘tell legend’ > yəhubtxʷ ‘recite

legend to someone’). On the whole, this class of verbs —to the extent that it is a coherent class

— is less consistent in terms of its derivational possibilities than the patient-oriented radicals.

It should also be noted that the division into patient-oriented and agent-oriented (or

unaccusative and unergative) radicals is by no means exhaustive, nor does it allow for hard-and-

fast predictions about which derivational affixes a particular radical will combine with. There

are, for instance, verbs of state (e.g., √hiiɬ ‘be happy’, √x&əc ‘be afraid’, √t’əbaʔ ‘have fallen in

water’) that do not pattern with the patient-oriented radicals in taking the internal causative.

Indeed, stative verbs expressing property concepts such as √luƛ’ ‘be old’ and √qʷiq’ʷ ‘be strong’

do not have transitive forms at all, while other property concept terms (e.g., haʔɬ ‘good’, hikʷ

‘big’) combine quite happily with the internal causative. Thus, while there are generalizations to

be made about a large number of Lushootseed radicals in terms of their combinatorial

possibilities, there is a very large class of unpredictable radicals with idiosyncratic derivational

patterns.

In addition to monovalent patient- and agent-oriented radicals, there are some inherently

bivalent verbal radicals, although these are few in number. The thirteen bivalent radicals found in

the textual corpus used for this grammar are given in Table 9:

º√ʔaladᶻ ‘care for ’

√qada ‘steal ’

√ʔuləx& ‘gather , forage for ’

√q’ʷuʔ ‘be together with ’

√čəbaʔ ‘be loaded down with ’

√pus ‘be hit by  (missile)’

√kʷukʷcut ‘cook ’

√šəɬ ‘make ’

√k’ʷič’ ‘butcher ’

√təxʷ ‘buy ’

√ɬəgʷɬ ‘leave ’

√xʷiʔxʷiʔ ‘hunt for , forage for ’

√ƛ’al ‘put  on’

Table 9: Bivalent radicals

All of these are bivalent intransitives with the exception of √ɬəgʷɬ ‘leave something’, which is a

true transitive verb in that it takes a direct object and can be passivized, as shown in (2):

(2) a. ʔuɬəgʷəɬ čəɬ ti kikəwič

ʔu–ɬəgʷəɬ čəɬ ti kikəwič

PFV–leave.behind 1PL.SUB SPEC ATTN–hunchback

‘we left Little Hunchback behind’

[LA Basket Ogress, line 121]

b. haˑy ɬəgʷəlb, x%ʷul’ ʔəsq’il ʔal tiʔəʔ q’il’bid

hay ɬəgʷɬ–b x%ʷul’ ʔəs–q’il ʔal tiʔəʔ q’il’bid

SCONJ leave.behind–PASS only STAT–aboard PR PROX canoe

‘and then [his corpse] was left, [it] was just aboard his canoe’

(Hess 1998: 92, lines 37–38)

Two of the radicals in the list, √šəɬ ‘make something’ and √təxʷ ‘buy something’, belong to the

class of verbs that take nominal predicate complements; they are bivalent because they

subcategorize for two arguments (a subject and a complement), but are intransitive in that the

complement is not a full NP or a true direct object (Section 8.2.5). The remainder of the radicals

in Table 9 subcategorize for oblique objects introduced by the preposition ʔə:

(3) a. qada čəxʷ ʔu ʔə tə sduukʷ

qada čəxʷ ʔu ʔə tə sduukʷ

steal 2SG.SUB INT PR NSPEC knife

‘did you steal the knife?’

(Bates, Hess & Hilbert 1994: 172)

b. ləsčəbaʔ ʔə tə hud

ləs–čəbaʔ ʔə tə hud

CONT–laden PR NSPEC wood

‘she’s shoulder-packing the wood’

(Bates, Hess & Hilbert 1994: 61)

With the exception of º√ʔaladᶻ ‘care for something’ and √kʷukʷcut ‘cook something’ (the latter

a borrowing based on English cook), these bivalent radicals are more or less productive bases for

derivation, although as a set they do not show any predictable combinatory patterns, other than

that none appears with the causative of activity (Section 2.1.2.4), which normally serves to

derive a bivalent intransitive stem from a patient-oriented radical. The only valency-increasing

affixes that appear with two of these radicals, √qada ‘steal something’ and √xʷiʔxʷiʔ ‘hunt for

something, forage for something’, are secondary suffixes (2.1.3); on the other hand, √ʔuləx&

‘gather something, forage for something’, √čəbaʔ ‘be loaded down with something’, √ƛ’al ‘put

something on’, √q’ʷuʔ ‘be together with something’, and √pus ‘be hit by something (missile)’

each take the internal causative (2.1.2.1 and at least one other valency-increaser. The nominal-

complement taking radicals, √šəɬ ‘make something’ and √təxʷ ‘buy something’, both combine

with the external causative -txʷ (2.1.2.2), and √šəɬ ‘make something’ takes the diminished

control causative -dxʷ (2.1.2.3) as well.

Because of the reliance of the Lushootseed lexicon on monovalent radicals, and because so

many of those radicals have stative or state-like meanings, the language has developed a prolific

system of derivational affixation. In total, there are fourteen derivational suffixes and one

derivational prefix that are used productively to derive new lexemes from verbal radicals. These

affixes fall roughly into two sets — valency-neutral affixes and valency-increasing affixes. The

valency-neutral affixes — dxʷ(s)- ‘contained’, -il ‘inchoative’, -b ‘middle’, -agʷil ‘autonomous

action’, -iluɬ ‘purposive’, and -áb ‘method’ — effect a variety of changes in the Actionsart or

semantic structure of the event expressed by the radical without affecting the number of event-

participants (semantic actants) or syntactic arguments. Valency-increasers, on the other hand,

add semantic and actants and increase the syntactic valency of their bases. In total there are ten

of these: -t ‘internal causative’, -txʷ ‘external causative’, -dxʷ ‘diminished control causative’, -b

‘causative middle’, -alikʷ ‘causative of activity’, -c/-s ‘allative applicative’, plus the secondary

suffixes (-yi- ‘dative applicative’, -bi- ‘middle applicative’, and the fossilized stem formatives

-di-/-i-) which are always combined with another valency-increaser. Related to the category of

valency-increasing affixes is the incorporative suffix -əɬ, which adds a predicate complement to

the valency of its base rather than an NP object. Each of these affixes will be discussed in turn in

the sections below, beginning with the valency-neutral affixes (Section 2.1.1) and the valency-

increasing (2.1.1.7) and secondary suffixes (2.1.3), followed by a discussion of the incorporative

(2.1.4) and an additional affix, the propriative, used to derive verbs from nominal bases (2.1.5).

A single radical can take one or more of both types of affix. A rough indication of the

combinatorial potential and relative ordering of these morphemes is given in Figure 1.

-1

dxʷ(s)- ‘CTD’

√R

(V)

-il ‘INCH’

-b ‘MD’

-agʷil ‘AUTO’

-alikʷ ‘ACT’

-iluɬ ‘PRPV’

-əɬ ‘INCRP’

-áb ‘MTHD’

-txʷ ‘ECS’

-dxʷ ‘DC’

-yi- ‘DAT’

-bi- ‘MAP’

-di-/-i- ‘SS’

-t ‘ICS’

-c/-s ‘ALTV’

-b ‘CSMD’

Figure 1: Affixal template of the Lushootseed verb stem

The numbered columns indicate relative position of each affix with respect to the radical (R) and

the other affixes; these are meant only to indicate linear precedence and don’t necessarily

correlate with a particular semantic type of affix, although the valency-neutral affixes are closer

to the stem, in Positions 1 and 2. Affixes that occur in the same column are not attested co-

occurring in the same stem. Although the template in Figure 1 assigns a specific position to each

of the derivational affixes, there is no evidence to date for some of the positions assigned to

individual affixes. For instance, there are no forms that show the allative applicative combining

with one of the secondary suffixes (Position 4), meaning that it could also have been placed in

Position 3.

There are also a few forms such as cilyialikʷ ‘dish something out for someone’ and

ƛ’iq’ačiʔyibtxʷ ‘make someone’s hands sticky’ where the affixes show a different order,

probably indicating a cyclical diachronic process where morphologically-complex stems become

lexicalized enough to be reanalyzed or treated as unanalyzed bases for subsequent derivation.

Not unsurprisingly, the least precise part of the template is that part nearest to the radical —

Positions 1 and 2. The suffixes -iluɬ ‘purposive’ and -əɬ ‘incorporative’, for instance, are not

attested following other derivational affixes, and so have been placed in either Position 1 or

Position 2. The suffix -áb ‘method’ is found following at least one stem, dᶻəqil ‘crawl’, which, at

least historically, contains the inchoative suffix -il; however, it may be that this is a fossilized

form as far as the morphology is concerned, and that -áb appears in Position 1 and is in fact

synchronically incompatible with other derivational suffixes. Likewise, forms that contain a

number of the possible combinations of Position 1 and 2 suffixes — as well as some of the

potential combinations of these with Position 3 affixes — are rare or unattested. Nevertheless,

the relative positions of the affixes shown in Figure 1 hold for the vast majority of wordforms,

and are probably an accurate depiction of the productive derivational processes of the synchronic

grammar.

2.1.1 Valency-neutral verbal affixes

The valency-neutral affixes are derivational affixes that attach to a verbal radical to derive a

new lexeme but which leaves its semantic and syntactic valency unchanged. In total there are

five of these, one prefix — dxʷ(s)- ‘contained’ — and five suffixes — -il ‘inchoative’, -b

‘middle’, -agʷil ‘autonomous action’, -iluɬ ‘purposive’, and -áb ‘method’.

There is also only one form, tupyib ‘pound something to prepare as food”, which places the causative middle after

the secondary suffix -yi-; however, this is the only attested form so far that combines a secondary suffix and -b,

which otherwise seem not to co-occur.

2.1.1.1 Contained dxʷ(s)-

The prefix dxʷ(s)- ‘contained [CTD]’ is one of the most frequent derivational prefixes in the

Lushootseed lexicon and is found as part of a great many verb stems. Its basic meaning is to

indicate that something is contained while undergoing the process expressed by its base:

(4) a. ʔuhəd čəd

ʔu–həd čəd

pfv–warm 1sg.sub

‘I got warmed up (after being outside)’

(Bates, Hess & Hilbert 1994: 107)

b. putəxʷ dxʷshəd tiʔəʔ č’ič’ič’ƛ’aʔ səshudčupyids

put=əxʷ dxʷs–həd tiʔəʔ č’i–č’i–č’ƛ’aʔ s=ʔəs–hud•čup–yi–d=s

PTCL=now CTD–warm PROX ATTN–ATTN–stone NM=burn•fire–DAT–ICS=3PO

‘the stones she had put in the fire for them were very hot now’

[DM Basket Ogress, line 67]

c. ɬudxʷshədiləxʷ tiʔəʔ č’ƛ’aʔ

ɬu=dxʷs–həd–il=əxʷ tiʔəʔ č’ƛ’aʔ

IRR=CTD–warm–INCH=now PROX stone

‘the stones will warm up in there (the fire)’

[LA Basket Ogress, line 66]

In forms such as dxʷshəd in (4b), the containment is fairly literal (in a fire pit); in other cases the

containment is more figurative, often psychological (e.g., dxʷx&ʷax&ʷaq’ʷacut ‘be troubled’,

dxʷsqʷacdxʷ ‘doubt someone’) or having to do with some inherent feature of an object

(dxʷx&ʷədᶻəb ‘be tart, be strong (coffee)’ from √x&ʷəc ‘be sharp’). Many verbs expressing actions

associated with water are formed with dxʷ(s)- as well — e.g., dxʷbəčəb ‘sink, drown’, dxʷčəcəb

‘be red (water)’, dxʷqələbil ‘be turbid (water)’, dxʷšəɬəb ‘look at something through water’.

In addition to being one of the more frequent derivational prefixes, dxʷ(s)- is one of the most

difficult to analyze, primarily because it has three separate allomorphs — [dxʷs-], [dxʷ-], and

[xʷ-] — which appear to be to a certain extent in free variation with one another: the form of the

affix chosen for use with a particular base varies both among dialects and amongst individual

speakers of the same dialect. The waters are muddied further by the fact that each of these

allomorphs is homophonous with another affix that must be treated, at least synchronically, as a

separate morpheme — specifically, [dxʷs-] is homophonous with dxʷs- ‘proclivitive’ (Section

2.2.1), [dxʷ-] is homophonous with the directional particle dxʷ (2.7.2), and [xʷ-] is

homophonous with xʷ- ‘container’ (2.2.5); dxʷ(s)- is also part of the expression of the

desiderative, discussed below in Section 8.5 (see Hess 1971 for a thorough comparison of these

affixes). Historically, at least some attestations of dxʷ(s)- are probably analyzable as dxʷ- plus

the s=nominalizer, and much of the modern-day variation between [dxʷs-] and [dxʷ-] may be the

result of the neutralization of what were diachronically different structures: as the two collapsed

into a single morpheme, dialects and speakers began to reanalyze what were once *dxʷ-s- forms

(these being, perhaps in some cases, clause-level nominalizations) as [dxʷ-], or vice versa. The

same diachronic processes may also account for the origins of some or all of the homophonous

affixes, although all of these possibilities await areful etymological analysis and reconstruction.

dxʷ(s)- interacts morphophonemically with the stative aspectual prefix ʔəs-:

(5) a. ʔudxʷyaƛ’ b. ʔəxʷyaƛ’

ʔu–dxʷ–yaƛ’ ʔəs–dxʷ–yaƛ’

PFV–CTD–dipped.out STAT–CTD–dipped.out

‘it got filled with water’ ‘it is filled with water’

(Hess 1998: 26)

The resultant form is found with all three allomorphs of dxʷ(s)-, as well as in the stative forms of

desiderative expressions (Section 8.5).

A number of stems formed with dxʷ(s)- are given in Table 10:

dxʷbəčəb ‘sink, drown’

(√bəč ‘be lying, be fallen from standing’)

cəb ‘be əd ‘sink ’

(√bəč ‘be lying, be fallen from standing’)

dxʷčəcəb ‘be red (water)’

(º√čc ‘red’; cf. x&ičəc ‘red’)

dxʷdigʷid ‘advise ’

(º√dikʷ ‘be advised’; cf. dikʷdxʷ

‘instruct ’)

dxʷp’əq’ ‘be filled with drifting things’

(√p’əq’ ‘drift’)

dxʷpədəb ‘have dust-covered water inside’

(√pəd ‘be dirty, be buried’)

dxʷqədid ‘cuckhold ’

(º√qəd ‘fornicate’; cf. qədəb ‘have illicit sex with ’)

dxʷqələbil ‘be turbid (water)’

(√qələb ‘(to)rain’)

dxʷshəd ‘be hot in container; be hot (weather)’

(√həd ‘be hot’)

dxʷshədil ‘heat up in container; get hot (weather)’

(from √həd ‘be hot’ + -il ‘inchoative’)

dxʷšab ‘dry out (container or body of water)’

(√šab ‘be dry’)

dxʷšəɬəb ‘look at  through water’

(√šuɬ ‘appear, be visible’)

dxʷt’iq’ʷəb ‘be murky (water)’

(º√t’iq’ʷ ‘be smoky/murky’; cf. t’iq’ʷalikʷ ‘smoke  (meat)’)

dxʷt’iq’ʷil ‘get murky (water)’

(from º√t’iq’ʷ ‘be smoky/murky’ + -il ‘inchoative’)

dxʷx&ʷax&ʷaq’ʷacut ‘be troubled’

(√x&ʷaq’ʷ ‘be worried, be preoccupied’)

dxʷx&ʷədᶻəb ‘be tart, be strong (coffee)’

(√x&ʷəc ‘be sharp’ +-b ‘middle’)

dxʷyaƛ’ ‘be filled with water’

(yaƛ’əb ‘carry water’ based on º√yaƛ’ ‘be dipped out’)

xʷtux&ʷtux&ʷalikʷ ‘(spider) stretches  (web)’

(º√tux&ʷ ‘be stretched’; cf. tux&ʷud ‘stretch ’)

xʷtux&ʷud ‘(spider) stretches  (web)’

(tux&ʷud ‘stretch ’)

Table 10: Stems formed with dxʷ(s)-

To this list could also be added less transparent verbal forms such as dxʷət’əb ‘be clear’,

dxʷbilcəb ‘fall on rump’, dxʷʔəhad ‘talk over’, dxʷsax&əbiləxʷ ‘beginning to get bitter’ (based

on sax& ‘scrape’), dxʷt’acəbiləxʷ ‘beginning to get bitter’, dxʷkʷədibəɬ ‘make off with someone

else’s wife or girlfriend’ (based on √kʷəd ‘be held, be taken’), dxʷk’ic’id ‘eviscerate something’,

dxʷsaqtxʷ ‘make motions as if to hit someone’, and dxʷsəlpaləwəpəd ‘use a fire-making drill’

(based on √səlp ‘spin’). dxʷ(s)- also appears in one or two nominal expressions — for example,

dxʷsacəb ‘rapids’, dxʷšulaʔkʷčup ‘drill for making fire’ (based on √šul ‘be pushed in’ + -čup

‘fire’) — none of which is completely analyzable. In addition, it is found in a pair of verbal

stems with an idiomatic meaning — dxʷsq’ʷəl ‘be hot (weather)’ and dxʷsq’ʷəlil ‘get hot

(weather)’ (from √q’ʷəl ‘be cooked, be ripe’) (cf. dxʷshəd ‘be hot in container; be hot (weather)’

and dxʷshədil ‘heat up in container; get hot (weather)’ in Table 10). There are also two forms —

dxʷsqʷacdxʷ ‘doubt someone’ and dxʷshaydxʷ ‘know something’ — that appear in the texts

with dxʷ(s)- but are also well-attested without the prefix (as qʷacdxʷ ‘doubt someone’ and

haydxʷ ‘know something’, respectively). There is no obvious distinction in meaning between the

plain and prefixed forms, although both verbs obviously have a certain inherent (psychological)

internality, making the prefix superfluous and allowing for the co-existence of two competing

synonymous or near-synonymous forms. Hess (1971) reports that in many cases forms of the

same stem with and without dxʷ(s)- are deemed to be synonymous by speakers.

One feature of dxʷ(s)- that has attracted the attention of researchers is the ways in which it

interacts with lexical suffixes (Section 2.1.6). Forms which combine dxʷ(s)- and a lexical suffix

fall into three categories. The first of these are transparent constructions in which both affixes

have their usual meanings, such as those in (6):

(6) a. dxʷc’agʷəliʔəd

dxʷ–c’agʷ–əliʔ–əd

CTD–washed•ear–ICS

‘she washed his ears out’

b. dxʷc’ibwild tə ɬaʔx%

dxʷ–c’ib•wil–d tə ɬaʔx%

CTD–lick•canoe–ICS NSPEC small.bowl

‘lick the pan’

In (6a), the lexical suffix -əliʔ ‘ear; side of head’ (a reduced form of -aladiʔ) has its usual

function, specifying the active zone of the action on the PATIENT’s body, while dxʷ(s)- (here,

[dxʷ-]) expresses the fact that the washing took place (or at least included) the inside of the

PATIENT’s ears. Similarly, dxʷ(s)- in (6b) indicates that the AGENT is to lick the inside of the pan,

which is further specified by the suffix -wil ‘canoe; narrow passage’ (a reduced form of -gʷiɬ

whose final consonant has been voiced by the following suffix). Similar expressions are found

throughout the corpus. This category of transparent dxʷ(s)- + lexical suffix constructions might

also include (or at least be the source of) one or two nominal expressions such as dxʷʔílalədiʔ

‘cheek’ (lit. ‘inside the edge (ʔil) of the side of the head’ and dxʷbaʔwiɬ ‘wide canoe’ (based on

baʔ ‘be wide’). As noted in Hess (1971), two lexical suffixes — -ačiʔ ‘hand’ and -šəd ‘lower

leg’ — never appear in conjunction with dxʷ(s)-, possibly because these bodyparts are already

highly specific, or because — as body extremities — they are peripheral rather than central to the

body and so are not compatible the notion of internality or containment.

The next category of dxʷ(s)- + lexical suffix constructions is the set of conventionalized

combinations which have a specialized meaning:

dxʷ- -qid ‘voice’

-qid ‘head’

dxʷ- -qs ‘nose’

-qs ‘point’

dxʷ- -ucid ‘language’

-ucid ‘mouth’

dxʷ- -us ‘face’ or ‘hair’

-us ‘top, upper’

Table 11: Conventionalized lexical suffix combinations with dxʷ(s)-

While the diachronic path for each of these combinations is fairly clear (the voice is contained

within the head, the nose is in the centre of the face, etc.), synchronically these pairs of affixes

have the function of ordinary lexical suffixes. All of the attestations of the prefix in this

combination in the present corpus are of the [dxʷ-] allomorph.

Finally, there are half a dozen lexical suffixes which are unattested without dxʷ(s)-:

dxʷ- -ap/-ah ‘bottom, rump’

dxʷ- -apsəb ‘neck, nape’

dxʷ- -igʷəd ‘inside, interior’

dxʷ- -wič ‘back’

dxʷ- -yax&ad ‘hinged door’

dxʷ- -yuq’ʷ ‘throat’

Table 12: Lexical suffixes requiring dxʷ(s)-

In these forms, dxʷ(s)- seems to be entirely bleached of its meaning and is simply part of the

lexical affix. All of the attestations of the prefix in this combination in the present corpus are also

of the [dxʷ-] allomorph.

2.1.1.2 Inchoative -il

One of the most frequent derivational affixes found on verbs is -il ‘inchoative [INCH]’ which,

when affixed to a monovalent radical expressing a state X, creates a monovalent stem expressing

the meaning ‘come into state X’:

(7) a. wəliʔ ƛ’usəsaʔliʔ ʔal tiʔiɬ ƛ’uq’il’bids sxʷiʔxʷiʔs əlgʷəʔ

wəliʔ ƛ’u–s–ʔəs–saʔliʔ ʔal tiʔiɬ ƛ’u–q’il’bid–s s=xʷiʔxʷiʔ=s əlgʷəʔ

visible HAB–STAT–two PR DIST HAB–canoe–3PO NM=hunt=3PO PL

‘the usual two hunting in their canoe [will] be visible’

(Hess 2006: 49, line 187)

More accurately, this affix could be characterized as a transmutative, which Mel’çuk (1993-2000: vol. 2, 349)

defines as a morpheme expressing the meaning ‘to begin to be [X]’. This is as opposed to his definition of the

inchoative, ‘to begin [X]’. In this sense, a transmutative is a particular case of an inchoative which in most

languages would apply to an adjective (state) whereas the typical inchoative applies to a verb (process). Given that

Lushootseed does not distinguish between these two word classes — and that there is no contrasting affix to -il that

correspond to the more typical “verbal” inchoative — the more general (and generally recognized) term has been

chosen here in line with most other work on Lushootseed (with the exception of Beck 1996) and other Salishan

languages.

b. ʔuwəliʔiləxʷ stab hiˑkʷ əw’ə x%pay’ac tiʔəʔ ləsaxʷəbtxʷ əlgʷəʔ

ʔu–wəliʔ–il=əxʷ stab hikʷ əw’ə x%pay’ac tiʔəʔ lə=saxʷəb–txʷ əlgʷəʔ

PFV–visible–INCH=now what big PTCL cedar PROX PROG=run–ECS PL

‘that which was the big cedar tree that ran off with them appeared’

(Hess 2006: 53, line 268)

A representative sample of verbs from synchronically analyzable bases is given in Table 13:

ʔacigʷədil ‘be inside of’

(from º√ʔac ‘centre’ + -igʷəd ‘body’)

ʔaʔil ‘come to be in a place’

(√ʔa ‘be there’)

ʔalil ‘come to the point where’

(√ʔal ‘on, at’)

ʔiʔabil ‘become wealthy’

(º√ʔiʔab ‘be wealthy’; cf. sʔiʔab ‘noble person’)

ʔudəgʷičil ‘get into middle of house’

(from º√ʔudəgʷ ‘centre’ + -ič ‘covering’)

bək’ʷil ‘get used up, be done exhaustively’

(√bək’ʷ ‘all, completely’)

biƛ’il ‘get crushed’

(º√biƛ’ ‘be smashed, be crumbled’; cf. biƛ’id ‘smash ’)

buusil ‘become four’

(√buus ‘four’)

čadᶻil ‘become hidden, hide self’

(√čac ‘be hidden’; cf. čactxʷ ‘hide ’)

čagʷil ‘get out to sea’

(√čaʔkʷ ‘seaward’)

čcil ‘turn red’

(º√čc ‘be red’; cf. x&ičəc ‘red’)

č’itil ‘draw near’

(√č’it ‘nearby’)

diʔil ‘go off a-ways’

(√diʔ ‘on the other side’)

dukʷil ‘become strange, become supernatural’

(√dukʷ ‘be anormal’)

gəqil ‘become dazzling; clear up [weather]’

(º√gəq ‘shining’; cf. gəqəb ‘[sun] shines’)

haʔɬil ‘become good’

(√haʔɬ ‘good’)

hədil ‘warm up’

(√həd ‘be warm’)

həliʔil ‘become well, heal’

(√həliʔ ‘be alive’)

higʷil ‘become noble’

(√hikʷ ‘big’)

hiqil ‘get pushed up’

(º√hiq ‘be pushed’; cf. hiqid ‘push ’)

huyil ‘become’

(√huy ‘be done, be made, be finished’)

ǰuʔil ‘have a good time, be happy’

(º√ǰuʔ ‘be glad’; cf. ǰuʔid ‘entertain ’)

k’ʷilil ‘peer’

(√k’ʷil ‘peek’)

laqil ‘become last’

(√laq ‘be last’)

ləliʔil ‘become different’

(√ləliʔ ‘be different’)

ləx&il ‘get light, get bright’

(√ləx& ‘be light, be bright’)

luƛ’il ‘grow old, grow up’

(√luƛ’ ‘be old’)

ɬalil ‘get out of fire’

(º√ɬal ‘be removed from fire’; cf. ɬalš ‘remove  from fire’)

ɬax&il ‘get dark’

(√ɬax& ‘be dark’)

ƛ’ubil ‘get better’

(√ƛ’ub ‘good, well’)

qahil ‘become a lot’

(√qah ‘be a lot’)

qʷatil ‘become laid out’

(√qʷat ‘be lying; snow falls’)

qʷəq’ʷil ‘turn white’

(º√qʷq’ʷ ‘be white’; cf. x&iqʷəq’ʷ ‘white’)

qʷiq’ʷil ‘become strong’

(√qʷiq’ʷ ‘be strong’)

q’ʷəlil ‘get hot’

(√q’ʷəl ‘be cooked, be ripe’)

saʔil ‘get in trouble’

(√saʔ ‘be bad’)

šqil ‘rise up’

(√šq ‘be high’)

This radical is grouped more properly with the adverbs, though it is included here because in terms of its

derivational properties it patterns with verbal radicals.

This radical is grouped more properly with the adverbs, though it is included here because in terms of its

derivational properties it patterns with verbal radicals.

See fn. 8 above.

This radical is grouped more properly with the adverbs, though it is included here because in terms of its

derivational properties it patterns with verbal radicals.

šuk’ʷil ‘turn grey’

(√šuk’ʷ ‘powder’)

tudəqil ‘become enslaved’

(º√tudəq ‘be enslaved’; cf. studəq ‘slave’)

t’agʷil ‘get on top’

(√t’agʷt ‘be on top’)

wəliʔil ‘become visible’

(√wəliʔ ‘be visible, be apparent’)

xʷiʔil ‘become non-existent’

(√xʷiʔ ‘no, not’)

Table 13: Stems formed with -il

In addition to these verbs, there are a great many stems that appear to be formed from the

combination of the inchoative and a verbal radical which is never attested without -il. These

include a number of verbs of motion (e.g., xʷit’il ‘fall, descend’, gʷəcil ‘wade’, ɬalil ‘go

ashore’), posture (gʷədil ‘sit down’, tudᶻil ‘bend forward’, tədᶻil ‘lie in bed’), and mental or

physical states (x&icil ‘get angry’, xʷak’ʷil ‘be tired’), as well as a considerable number of

miscellaneous stems (hiwil ‘go ahead’, ƛ’uil ‘become thin’). While forms such as these are

synchronically unanalyzable, they are almost certainly derived diachronically from inchoative

forms.

Stems formed with the inchoative are frequently used as bases for further derivation through

the addition of valency-increasing affixes (Section 2.1.1.6). Combinations of -il and the internal

causative -t include šuk’ʷild ‘grey something’ (from šuk’ʷil ‘turn grey’ based on √šuk’ʷ

‘powder’) and ʔaʔild ‘put somewhere’ (from ʔaʔil ‘come to be in a place’ based on √ʔa ‘be in a

place’). The stem q’ʷəliltxʷ ‘warm something up’ (√q’ʷəl ‘be cooked, be ripe’, via q’ʷəlil ‘get

hot’) is based on the external causative, -txʷ, as is qəliltxʷ ‘stop, warn’ (from √qəl ‘be stopped’),

although there is no intermediary form, *qəlil. The diminished control causative, -dxʷ, combines

with the inchoative in forms such as dukʷildxʷ ‘distrust someone’ (from dukʷil ‘become strange’

based on √dukʷ ‘be a-normal’), bək’ʷildxʷ ‘managed to finish something’ (from bək’ʷil ‘be

finished off’ based on √bək’ʷ ‘all, completely’), and qʷiq’ʷildxʷ ‘strengthen something’ (from

qʷiq’ʷil ‘become strong’ based on √qʷiq’ʷ ‘strong’). In addition, there are a variety of forms

This radical is grouped more properly with the adverbs, though it is included here because in terms of its

derivational properties it patterns with verbal radicals.

such as p’aʔxʷəbiltxʷ ‘disfavour someone’ which appear to contain both the inchoative and one

of the valency-increasers (in this case, -txʷ), but which are not synchronically analyzable.

Stems formed with -il also appear to serve as bases for the middle -b (Section 2.1.1.3), as in

t’asbil ‘pay for something’ (from º√t’as), ʔukʷukʷilb ‘get involved in play’ (√ʔukʷukʷ ‘play’),

and qadaʔilb ‘be stealing something, go around stealing something’ (√qadaʔ ‘steal something’).

All of these stems have activity reading typical of middle forms in Lushootseed and other

languages, and in none of these cases is there an intermediate inchoative form (i.e., *t’asəb,

*ʔukʷukʷil, or *qadaʔil).

With two fairly common radicals, √pus ‘be hit with something (missile), have something

thrown at one’ and √t’uc’ ‘be shot, be shot at’, the inchoative has a rather idiosyncratic effect on

the government pattern of its base, as in (8):

(8) a. ʔupus čəd ʔə ti baseball

ʔu–pus čəd ʔə ti baseball

PFV–be.thrown.at 1SG.SUB PR SPEC baseball

‘I got hit by a baseball’

b. ʔupusil čəd ʔə ti ʔəsbulux%ʷilc

ʔu–pus–il čəd ʔə ti ʔəs–bulux%ʷ–ilc

PFV–be.thrown.at–INCH 1SG.SUB PR SPEC STAT–be.round–round.thing

‘I threw the ball’

(Bates, Hess & Hilbert 1994: 164)

The first example shows the bivalent intransitive radical √pus ‘be hit with something (missile)’,

which takes as its subject the expression of the GOAL and as its oblique object the expression of

the INSTRUMENT. When -il is added to this radical, the resulting stems takes as its subject an

AGENT (which is not expressible in clauses formed on the bare radical). The GOAL is not

expressible with this stem, and the INSTRUMENT continues to be realized as an oblique object. A

similar pattern is seen with √t’uc’ ‘be shot, be shot at’:

(9) a. ʔut’uc’ čəd

ʔu–t’uc’ čəd

PFV–get.shot 1SG.SUB

‘I got shot’

(Bates, Hess & Hilbert 1994: 241)

b. ʔut’uc’il čəd ʔə tiʔiɬ t’isəd

ʔu–t’uc’–il čəd ʔə tiʔiɬ t’isəd

PFV–get.shot–INCH 1SG.SUB PR DIST arrow

‘I shot an arrow’

(Hess & Hilbert 1976: II, 131, ex. 21)

Unlike √pus, √t’uc’ seems to be monovalent, or at any rate unattested with an oblique object

expressing an INSTRUMENT. However, the inchoative form of the radical shows the same

government pattern as the inchoative of √pus, and shows the same replacement of a subject

expressing a GOAL with the bare radical by a subject expressing an AGENT in the inchoative. Both

of these forms can then be transitivized with the diminished control suffix -dxʷ:

(10) a. ʔupusildxʷ čəd ti ʔəsbulux%ʷilc

ʔu–pus–il–dxʷ čəd tə ʔəs–bulux%ʷ•ilc

PFV–be.thrown.at–INCH–DC 1SG.SUB NSPEC STAT–be.round•round.thing

‘I threw a ball’

(Bates, Hess & Hilbert 1994: 164)

b. ʔut’uc’ildxʷ čəd tiʔiɬ t’isəd

ʔu–t’uc’–il–dxʷ čəd tiʔiɬ t’isəd

PFV–get.shot–INCH–DC 1SG.SUB DIST arrow

‘I shot an arrow’

(Hess & Hilbert 1976: II, 130, ex. 16)

This pattern is not found with any other stems — or with any other affixes — in the language.

In addition to its role as a verb-formative, -il is also a part of a number of interesting

morphosyntactic constructions. One of these involves the combination of the inchoative with

numerals and the classifier -aɬ, used for counting time, to form expressions such as ɬixʷaɬil

‘be/happen three times’, buusaɬil ‘be/happen four times’. The inchoative also combines with the

Note that a variant of this sentence, ʔut’uc’il čəd tiʔiɬ t’isəd, in which the INSTRUMENT is expressed as a direct

object, appears in the Lushootseed Dictionary (p. 242). This form, attested only from a single consultant, is now

considered suspect by Hess as it was elicited under duress via translation/elicitation. This government pattern of the

stem has not turned up in spontaneous utterances, in examples provided by other speakers, or in texts.

negative incorporative predicate xʷəɬ ‘not have something’ to form the predicate xʷəɬil ‘run out

of something’:

(11) a. ʔəsxʷəɬ talə čəɬ

ʔəs–xʷəɬ talə čəɬ

STAT–not.have money 1PL.SUB

‘we don’t have money’

(Bates, Hess & Hilbert 1994: 252)

b. ʔuxʷəɬiləxʷ əlgʷəʔ sʔəɬəd

ʔu–xʷəɬ–il=əxʷ əlgʷəʔ sʔəɬəd

PFV–not.have–INCH=now PL food

‘they ran out of food’

[AW Basket Ogress, line 17]

These predicate-complement structures are discussed further in Section 8.2.5.

2.1.1.3 Middle -b

The valency-neutral middle marker -b ‘middle [MD]’ is used to form verb stems denoting

activities, processes, and other event-types identified by Kemmer (1993) as belonging to the

semantic domain of the middle in a wide range of languages. In its most productive use, the

valency-neutral middle is added to a verbal radical to form a monovalent intransitive stem as in

(12), where the middle marker has been added to the stative radical √luƛ’ ‘be old’ (12a) to create

a process verb luƛ’əb ‘age, grow old’ (12b):

(12) a. ƛ’al’ čəd bəluƛ’ x%ʷəl’ab ʔə dəgʷiʔ

ƛ’al’ čəd bə=luƛ’ x%ʷul’ab ʔə dəgʷiʔ

also 1SG.SUB ADD=old just.like PR you

‘I’m as old as you’

b. ƛ’al’ čəxʷ ɬubəluƛ’əb x%ʷul’ab ʔə ʔəca

ƛ’al’ čəxʷ ɬu=bə=luƛ’–əb x%ʷul’ab ʔə ʔəca

also 1SG.SUB IRR=ADD=old–MD just.like PR you

‘you too are going to get old like me’

(Bates, Hess & Hilbert 1994: 139)

As in this example, the middle suffix is most regularly found associated with verbal radicals. A

number of middle forms of this type are given in Table 14:

ʔaʔəb ‘be in a certain place’

(√ʔa ‘be there’)

baqʷuʔb ‘snow’

(√baqʷuʔ ‘be snow-covered’)

ckʷab ‘be taut’

(√cikʷ ‘be straight, be tautened’)

čagʷəb ‘be at sea’

(√čaʔkʷ ‘seaward’)

dxʷbəčəb ‘sink’

(√bəč ‘be lying, be fallen from standing’)

dᶻaƛ’əb ‘get confused’

(√dᶻaƛ’ ‘be confused’)

gəqəb ‘[sun] shines’

(º√gəq ‘shining’; cf. gəqil ‘clear up [weather]’)

haʔləb ‘be nice [weather]’

(√haʔɬ ‘good’)

hədʔiw’b ‘go inside’

(√hədʔiw’ ‘be inside a house’)

k’ʷalč’əb ‘bend self backwards’

(º√k’ʷalč’ ‘be bent backwards’; cf. k’ʷalč’əd ‘bend  backwards’)

luƛ’əb ‘age’

(√luƛ’ ‘be old’)

p’iləb ‘go flat’

(√p’il ‘be flat’)

p’q’adᶻəb ‘be rotting [log]’

(√p’q’ac ‘rotten log’)

qʷcab ‘slip’

(√qʷc ‘slide, slip’)

šabəb ‘dry out’

(√šab ‘be dry’)

yaƛ’əb ‘carry water’

(º√yaƛ’ ‘be dry’; cf. yaƛ’ad ‘scoop up  (water)’)

yəcəb ‘report on ’

(º√yəc ‘report on ’; cf. yəcəd ‘report ’)

Table 14: Monovalent intransitive stems formed with -b

These verb stems express events of a range of semantic types including states (ckʷab ‘be taut’,

čagʷəb ‘be at sea’), processes (luƛ’əb ‘age’, p’iləb ‘go flat’), activities (yaƛ’əb ‘carry water’),

non-translational motion (dxʷbəčəb ‘sink’, k’ʷalč’əb ‘bend self backwards’), and meteorological

expressions (baqʷuʔb ‘snow’, gəqəb ‘[sun] shines’, haʔləb ‘be nice [weather]’). In addition to

the analyzable forms shown in Table 14, there is a very large number of monovalent verbs that

end in -b whose radicals are otherwise unattested (or are attested only in forms with -b). These

include expressions of properties (e.g., qʷagʷəb ‘be sweet’, č’apəb ‘be sour’, ƛ’aɬəb ‘be salty’,

sadᶻəb ‘be tall’), body actions and non-translational motion (č’ədəb ‘shiver’, saxʷəb ‘jump,

sprint’), and certain meteorological expressions (qəlb ‘rain’, pahəd ‘be hazy’).

In several forms in Table 14, the middle marker derives verbs that denote an actor being in or

coming into the state denoted by the radical. This use seems to overlap with the inchoative -il

(Section 2.1.1.2), although there are a few pairs of stems formed from the same radical such as

haʔləb ‘be nice [weather]’ vs. haʔlil ‘turn nice [weather]’ and luƛ’əb ‘age’ vs. luƛ’il ‘grow up’.

Judging from the contrast between these pairs, the -b forms seem to refer more to the state and

the fact that the actor is in that state, whereas the -il forms are more telic and focus on the

See fn. 13 above.

achievement of the state denoted by the radical. However, there are relatively few such pairs and

most stative radicals that have inchoative or inchoative-like forms take either one or the other of

the two affixes; how predictable the selection of affix is from the semantics of the radical is a

topic for future investigation.

As revealed by the set of forms in Table 14, the valency-neutral middle -b has two regular

allomorphs, [-b] and [-əb], the former following vowels and approximants, and the latter

following obstruents. Even in the small sample shown here there are a few idiosyncratic forms.

The first of these is ʔaʔəb ‘be in a certain place’ from √ʔa ‘be there’; however, some speakers

use the form √ʔaʔ ‘be there’ instead of √ʔa, indicating that ʔaʔəb was probably a regular form for

all speakers in the not-too-distant past. Two of the stems in Table 14 — ckʷab ‘be taut’ (from

√cikʷ ‘be straight, be tautened’) and qʷcab ‘slip’ (from √qʷc ‘slide, slip’) — have an /a/

following the radical. This element is identified by Hess (1967a: 34) and Hess & Hilbert (1976:

II, 152) as a stem formative, although its semantic contribution to the derived form remains

obscure (see Section 2.1.6 below for further discussion).

A second class of middle forms involves the combination of the middle marker, a radical,

and a lexical suffix (Section 2.1.6). Those forms in which the lexical suffix has a literal meaning

have parallel transitive expressions with an overt NP argument:

(13) a. ʔuc’agʷad tiʔiɬ čaləs

ʔu–c’agʷa–d tiʔiɬ čaləs–s

PFV–wash–ICS DIST hand–3PO

‘s/he washed her/his hands’

(Hess 1998: 16, ex. c)

b. ʔuc’agʷačiʔb

ʔu–c’agʷ–ačiʔ–b

PFV–wash–hand–MD

‘s/he washed her/his hands’

(Hess 1998: 16, ex. b)

The bodypart expressed by the lexical suffix in such forms belongs to the AGENT/subject of the

sentence. Verbs derived in this way are monovalent and intransitive. While most such forms are

quite transparent, denoting an action taken by the AGENT/subject of the sentence involving a part

of its own body, a number of verbs of this type (e.g., ɬiʔɬda(hə)b ‘troll’ from √ɬid ‘be tied’ + -ah

‘bottom’, xʷəbaladiʔb ‘toss head side to side’ from º√xʷəb ‘be thrown’ + -adiʔ ‘ear’) have

figurative meanings. Middles formed with lexical suffixes constitute a very large class of verb

stems; a sample of these is given in Table 15:

ʔabačiʔəb ‘extend hands’

(º√ʔab ‘be extended’ + -ačiʔ ‘hand’)

ʔabšədəb ‘extend legs’

(º√ʔab ‘be extended’ + -šəd ‘leg’)

c’agʷačiʔəb ‘wash hands’

(√c’aʔkʷ ‘be washed’ + -ačiʔ ‘hand’)

dxʷqʷalusəb ‘paint face’

(√qʷal ‘be marked, be painted’ + -us ‘face’)

dᶻakʷaab ‘wag tail’

(√dᶻakʷ ‘be shaky’ + -ah ‘bottom’)

hədačiʔəb ‘warm up hands’

(√həd ‘be warm’ + -ačiʔ ‘hand’)

ɬiʔɬda(hə)b ‘troll’

(√ɬid ‘be tied’ + -ah ‘bottom’)

ƛ’ac’ahəb ‘cinch up belt’

(º√ƛ’ac’ ‘be cinched up’ + -ah ‘bottom’)

ƛ’iq’ačiʔbtxʷ ‘give sticky hands’

(√ƛ’iq’ ‘be sticky’ + -ačiʔ ‘hand’)

šəqlax&adəb ‘raise arms’

(√šq ‘be high’ + -l-ax&ad ‘arm’)

təq’aladiʔəb ‘slap sides of head’

(º√təq’ ‘be slapped’ + -al-adiʔ ‘ear’)

x&əqšadəb ‘wrap legs’

(º√x&q ‘be wrapped, be tied’ + -šad ‘leg’)

xʷəbaladiʔəb ‘toss head side to side’

(º√xʷəb ‘be thrown’ + -al-adiʔ ‘ear’)

Table 15: Middle forms with lexical suffixes

As the lexical suffixes are uniformly consonant-final, these forms require the [-əb] allomorph of

-b; however, in allegro speech forms such as c’agʷačiʔəb ‘wash hands’ and xʷəbaladiʔəb ‘toss

head side to side’ have slightly reduced forms — c’agʷačib (or c’agʷačiʔb) and xʷəbaladib (or

xʷəbaladiʔb), respectively. These forms, rather than the full forms, frequently appear in texts,

reflecting the storyteller’s actual pronunciation.

In addition to combining with verbal radicals, -b is also found associated with a few nominal

stems, given in Table 16:

bədaʔəb ‘have  (child)’

(√bədaʔ ‘offspring’)

bəlalwəb ‘tag along’

(√bəlalwəʔ ‘navel’)

bəšč’adəb ‘have lice’

(√bšč’ad ‘louse’)

dᶻulčuʔb ‘get bigger (waves)’

(√dᶻulčuʔ ‘wave (water)’)

kəpuub ‘wear coat’

(√kəpu ‘coat’)

p’q’adᶻəb ‘be rotting [log]’

(√p’q’ac ‘rotten log’)

ʔuq’ʷaʔəb ‘get younger sibling’

(√suq’ʷaʔ ‘younger sibling’)

Table 16: Middle forms based on nouns

The semantic effect of the middle suffix in these forms is rather idiosyncratic, ranging from a

semantic shift fairly typical of middles — ‘N’ > ‘have/use N’ (e.g., √bšč’ad ‘louse’ > bəšč’adəb

‘have lice’, √kəpu ‘coat’ > kəpuub ‘wear coat’, √suq’ʷaʔ ‘younger sibling’ > ʔuq’ʷaʔəb ‘get

younger sibling’) — to entirely idiomatic shifts (√bəlalwəʔ ‘navel’ > bəlalwəb ‘tag along’). In

one case, bədaʔəb ‘have something (child)’, the application of the middle seems to have created

a bivalent intransitive verb, a syntactic effect more commonly associated with the causative

middle -b (Section 2.1.2.4) (although in other respects this form is by no means a causative).

It should be noted here that the causative middle, -b, is treated in this grammar as a separate

morpheme from the valency-neutral middle suffix, based both on the distinct morphophonemics

of the two affixes, and their very different semantic and syntactic effects on their bases. As noted

above, the valency-neutral middle attaches directly to stems that take a harmonic vowel in the

presence of the internal causative -t (Section 2.1.2.1), while the causative middle triggers the

epenthesis of the harmonic vowel. Whereas the valency-neutral middle in most cases has no

effect on the valency of its base, the causative middle adds an AGENT, realized as a syntactic

subject, creating a bivalent intransitive stem that takes the former subject of the radical as an

oblique object. The causative middle also adds an element of the AGENT acting in its own interest

which is lacking in the semantics of stems formed with the valency-neutral middle, although —

as shown by Kemmer (1993) — the notion of an AGENT acting on itself or in its own interests is

also a cross-linguistically typical part of middle semantics. The two suffixes are, of course,

almost certainly cognates, as witnessed by the same range of uses and behaviours shown by the

cognate suffixes (-m in most cases) in other Salishan languages. The fact that the middle has split

into what are essentially two different morphemes from an analytical point of view is not

surprising, given the erratic behaviour that has been documented for middle-markers in a wide

range of languages, including familiar cases such as the Spanish se or the Russian -sʲa (Kemmer

1993). Nevertheless, in spite of their common diachronic origins, it is easier descriptively to treat

the two -b suffixes in Lushootseed as separate morphemes.

The same is true of the passive suffix (Section 6.2), which in Lushootseed (and again, in

many other Salishan languages) is homophonous with the valency-neutral middle-marker and the

causative middle. That the valency-neutral middle and passive suffixes are separate morphemes

is shown by their very different positions in the verbal complex and by their potential for co-

occurrence, as in (14):

(14) tuyəcəbtub čəd ʔə tiʔiɬ tudyəl’yəlab

tu=yəc–əb–txʷ–b čəd ʔə tiʔiɬ tu=d–yəl’–yəlab

PAST=report–MD–ECS–PASS 1SG.SUB PR DIST PAST=1SG.PO–DSTR–parent

‘it was told to me by my parents’

(Hess 1995: 140, line 3)

The distinction between the causative middle and the passive is, perhaps, less obvious as both

appear at the end of the verbal complex and are (outside of a few lexicalized forms) never

followed by other derivational affixes; however, the semantic and syntactic effects of the two

affixes are quite distinct. The causative middle has a clearly derivational function, deriving

bivalent intransitive stems from monovalent radicals, whereas the passive appears to be purely

inflectional and serves only to “shuffle” the diathesis of a verb, promoting a direct object to

subject and demoting a subject to an agentive complement. The parallels between passives and

middles are well-known, as are their functional overlaps in languages like Spanish, and so the

possible diachronic origin of the Lushootseed passive marker from a middle is not particularly

exotic. A potential semantic motivation for this diachronic path is discussed in Beck (1996).

2.1.1.4 Autonomous action -agʷil

The suffix -agʷil ‘autonomous action [AUTO]’ attaches to intransitive radicals describing

states to form intransitive verbs describing actions deliberately undertaken by the syntactic

subject leading to the subject being in that state.

(15) a. ʔubəč čəd

ʔu–bəč čəd

PFV–lying 1SG.SUB

‘I fell down (from a standing position)’

(Bates, Hess & Hilbert 1994: 35)

b. gʷəl ɬubəbəčagʷil

gʷəl ɬu=bə=bəč–agʷil

SCONJ IRR=ADD=lying–AUTO

‘he would go and lie down again’

(Hess 1998: 95, line 124)

This affix has two forms, the full form [-agʷil] and a reduced form [-awil] associated with Type

II attenuative reduplication (Section 5.2) — e.g., qʷcagʷil ‘slide’ vs. qʷiqʷcawil ‘go sledding or

skating’.

A sample of -agʷil forms is given in Table 17:

bəčagʷil ‘lie down’

(√bəč ‘be lying, be fallen from standing’)

dxʷbəčəbagʷil ‘go under water’

(dxʷbəčəb ‘sink’ from √bəč ‘be lying, be fallen from standing’)

gʷəx&agʷil ‘untie oneself’

(√gʷəx& ‘be untied’)

ƛ’iqagʷil ‘come out of’

(√ƛ’iq ‘emerge’)

ƛ’uq’ʷagʷil ‘cram self into small space’

(º√ƛ’uq’ʷ ‘be crammed in’; cf. ƛ’uq’ʷud ‘plug  in’)

q’ilagʷil ‘climb aboard’

(√q’il ‘be aboard’)

šulagʷil ‘enter cramped space’

(√šul ‘be in, be under’)

t’əbaʔagʷil ‘jump overboard’

(√t’əbaʔ ‘have fallen in water’)

xʷəbagʷil ‘throw oneself’

(º√xʷəb ‘be thrown’; cf. xʷəbed ‘toss ’)

xʷt’agʷil ‘climb down’

(º√xʷt’ ‘be fallen, be descended’; cf. xʷt’ad ‘take  down’)

x&qagʷil ‘tie oneself down’

(º√x&q ‘be wrapped, be tied’; cf. x&əqəd ‘wrap ’)

Table 17: Stems formed with -agʷil

The majority of these forms are verbs of non-translational motion formed from radicals

describing positions or physical configurations, the suffix adding the notion of the actor

deliberately taking up the position or configuration. This is particularly obvious in those

examples based on radicals such as √bəč ‘be lying, be fallen from standing’ and √t’əbaʔ ‘fall in

water’ which on their own lack any notion of agency:

This stem is also attested with the diminished control suffix, -dxʷ (Section 2.1.2.3), as šulagʷildxʷ ‘manage to get

into a cramped space’.

(16) a. ʔubəč čəd

ʔu–bəč čəd

PFV–fallen 1SG.SUB

‘I fell down’

(Bates, Hess & Hilbert 1994: 35)

b. gʷəl ɬubəbəčagʷil

gʷəl ɬu=bə=bəč–agʷil

then IRR=ADD=fallen–AUTO

‘then he would lie down again’

(Hess 1998: 95, line 124)

c. tucuuc čəd gʷəxʷiʔəs gʷəsuk’ʷit’s dxʷʔal tə stuləkʷ, gʷəl gʷət’əbt’əbaʔ

tu=cut–c čəd gʷə=xʷiʔ=əs gʷə=s=ʔu–k’ʷit’=s

PAST=say–ALTV 1SG.SUB SBJ=NEG=3SBJ SBJ=NM=PFV–go.down.to.water=3PO

dxʷ–ʔal tə stuləkʷ gʷəl gʷə=t’əb–t’əbaʔ

CNTRPT–at NSPEC river then SBJ=DSTR–fall.in.water

‘I told him not to go down the river, he might fall in’

(Bates, Hess & Hilbert 1994: 236)

d. gʷəl x%ʷul’ čəd gʷət’əbaʔagʷil čəda gʷəʔux%ʷ, gʷəʔusil, gʷət’ičib

gʷəl x%ʷul’ čəd gʷə=t’əbaʔ–agʷil čəda gʷə=ʔux%ʷ

then only 1SG.SUB SBJ=fall.in.water–AUTO 1SG.COORD SBJ=go

gʷə=ʔusil gʷə=t’ičib

SBJ=dive SBJ=swim

‘then I would just jump in the water, I’d go and dive and swim [away]’

[ML, Mink and Tutyika, line 32]

In each of these pairs of sentences, the first illustrates the use of the bare radical, which describes

an inadvertent, non-agentive action.

The second sentence of each pair shows the same radical

with -agʷil, denoting a deliberate, autonomous action on the part of the subject.

Although with most stems -agʷil does not effect any substantial change in the semantic role

assigned to the subject of the verb, in a few stems the suffix does in fact change the PATIENT role

assigned by the bare radical to that of an AGENT. In Table 17, there are three of these, a pair of

polar opposites, gʷəx&agʷil ‘untie oneself’ and x&qagʷil ‘tie oneself down’, and the verb

xʷəbagʷil ‘throw oneself’. Verbs like gʷəx&agʷil ‘untie oneself’ and x&qagʷil ‘tie oneself down’

In its stative aspectual form, bəč can have the meaning ‘be lying down’ and, with animate subjects, is ambiguous

as to whether the subject lay down deliberately or fell.

are based on radicals with a single argument that assign a patient-like semantic role; when used

with -agʷil, however, the semantic role assigned to the single argument of the verb is that of

AGENT, as in (17):

(17) gʷəl ɬubəx%qagʷil ʔal tiʔiɬ skəkiʔ

gʷəl ɬu=bə–x%q–agʷil ʔal tiʔiɬ skəkiʔ

SCONJ IRR=ADD–wrap–AUTO at DIST cradleboard

‘and again he will tie himself into the cradleboard’

(Hess 2006: 41, line 468)

Here the subject (Coyote disguised as a baby) deliberately ties himself into a cradleboard, an

action usually performed by a parent for (ideally) a passive child. The subject of the verb here is,

of course, also the PATIENT of the action, making this a semantically reflexive type of expression.

We see the same pattern in the form xʷəbagʷil ‘throw oneself’ (from º√xʷəb ‘be thrown’)

which describes an actor deliberately doing something that is normally done by an external

agent:

(18) tiləbəxʷ ʔuxʷəbagʷiləxʷ dxʷʔal tə qʷuʔ tiʔiɬ cədiɬ sup’qs

tiləb=əxʷ ʔu–xʷəb–agʷil–əxʷ dxʷ–ʔal tə qʷuʔ tiʔiɬ cədiɬ sup’qs

suddenly=now PFV–thrown–AUTO–now CNTRPT–at NSPEC water DIST it hair.seal

‘suddenly the hair seal threw itself into the water’

(Hess 2006: 50, line 213)

As in the previous example, the presence of the autonomous suffix here indicates that the subject

— in this case, the hair seal — deliberately does something (launches itself into the air) which

might normally be done by an external AGENT. In spite of the shift in semantic role accomplished

by -agʷil in such forms, the consistent effect of the affix across all of the stems it forms is to

heighten the agentivity of the actor.

2.1.1.5 Purposive -iluɬ

The suffix -iluɬ ‘purposive [PRPV]’ is a relatively infrequent affix which combines with a

stem expressing an action X which is potentially an activity and creates a verb meaning ‘go for

the purpose of doing X’:

(19) a. hay gʷəl ɬuxʷiʔxʷiʔ ʔə kʷi stab

hay gʷəl ɬu=xʷiʔxʷiʔ ʔə kʷi stab

SCONJ then IRR=forage PR REM what

‘and then he would forage for something’

(Hess 2006: 21, line 233)

b. bəxʷiʔxʷiʔiluɬ əlgʷəʔ

bə=xʷiʔxʷiʔ–iluɬ əlgʷəʔ

ADD=forage–PRPV PL

‘again they went to gather food’

(Hess 2006: 39, line 418)

-iluɬ forms are also bases for subsequent derivation:

(20) a. ʔux%ʷcəb ʔə tiʔiɬ stubš tsiʔiɬ sɬadəyʔ dxʷʔal kʷi gʷəsʔəɬdiluɬs əlgʷəʔ

ʔux%ʷ–c–b ʔə tiʔiɬ stubš tsiʔiɬ sɬadəyʔ dxʷ–ʔal kʷi

go–ALTV–PASS PR DIST man DIST woman CNTRPT–at REM

gʷə=s=ʔəɬd–iluɬ=s əlgʷəʔ

SBJ=NM=feed.on–PRPV=3PO PL

‘the man went after the woman to take her to lunch’

b. c’k’ʷaqid ʔuʔəɬdiluɬbitubuɬ

c’k’ʷaqid ʔu–ʔəɬəd–iluɬ–bi–t–ubuɬ

always PFV–eat–PRPV–MAP–ICS–1PL.OBJ

‘he always comes and eats off of us’

(Bates, Hess & Hilbert 1994: 11)

The analyzable -iluɬ forms attested in texts and listed in the entry for -iluɬ in the Lushootseed

Dictionary, as well as a few others, are given in Table 18:

ʔəɬdiluɬ ‘go out to eat’

(from ʔəɬəd ‘dine on ’)

c’əbəbiluɬ ‘go berry-picking’

(from c’əbəb ‘pick [berry]’ based on º√c’əb ‘clear land’)

dᶻubalikʷiluɬ ‘go to a dance’

(from dᶻubalikʷ ‘dance’ based on º√dᶻub ‘be kicked’)

gʷədiluɬ ‘go there to sit down’

(from gʷədil ‘sit down’ based on √gʷəd ‘down’)

qʷuʔqʷaʔiluɬ ‘go for a drink’

(√qʷuʔqʷaʔ ‘have a drink’)

q’əlbiluɬ ‘go camping’

(√q’əlb ‘camp out’)

t’iwiɬiluɬ ‘go to church’

(º√t’iwiɬ ‘practice religion’; cf. st’iwiɬ ‘religion’)

təpililuɬ ‘go salmon-fishing’

(from təpil ‘spear salmon’ based on √tp ‘be stabbed’)

xʷiʔxʷiʔiluɬ ‘go hunting/foraging’

(√xʷiʔxʷiʔ ‘hunt for , forage for ’)

yəy’duʔiluɬ ‘go for a swing’

(√yəy’duʔ ‘swing in a swing’)

Table 18: Stems formed with -iluɬ

In addition to these, the dictionary gives an unanalyzable form šəhabiluɬ ‘go to remove salmon

from trap’. Although most of the bases for -iluɬ in Table 18 are monovalent and give rise to

monovalent verbs, there are two bivalent intransitive bases — c’əbəb ‘pick something [berry]’

and √xʷiʔxʷiʔ ‘hunt for something, forage for something’. None of the contextualized attestations

of the -iluɬ forms derived from these bases have objects; however, it is not clear if this is an

accident of the contexts in which the verbs are used, or if it is because -iluɬ derivations are

necessarily monovalent intransitive stems.

2.1.1.6 Method -áb

The suffix -ab ‘method [MTHD]’ is a rather infrequent affix attached to nominal and verbal

roots to create verb stems that express the means or method of achieving something:

(21) a. čələsab tə səʔibəšs

čələs–ab tə s=lə=ʔibəš=s

hand–MTHD NSPEC NM=PROG=walk=3PO

‘he is walking on his hands’

b. dᶻəqilab ti dsuɬəgʷəɬ

dᶻəqil–ab ti d=s=ʔu–ɬəgʷəɬ

crawl–MTHD SPEC 1SG.PO=NM=PFV–leave

‘I left it by crawling’

(Hess & Hilbert 1976: II, 154)

c. sx%əyusab čəd ʔə tə dsəskiis

sx%əyus–ab čəd ʔə tə d=s=ʔəs–kiis

head–MTHD 1SG.SUB PR NSPEC 1SG.PO=NM=STAT–stand

‘I am standing on my head’

(Hess & Hilbert 1976: II, 154, fn. 4)

As seen in the form čələsáb (from √čaləs ‘hand’) in (21a), the method suffix in most cases

attracts stress and causes the reduction of vowels in the root to schwa. This process is blocked in

forms such as dᶻəqilab in (21b) that contain (or were formed historically with) the inchoative -il

(Section 2.1.1.2). The process also does not apply in many lexically-specified stems such as

sx&əyusab, seen in (21c). As seen in these examples, when affixed to a noun meaning ‘N’, -áb

creates a verb meaning ‘do X using N’ (21a); when affixed to a verb meaning ‘V’, it creates a

verb meaning ‘V by means of X’ (21b). In either case, the derived form is a bivalent intransitive

verb and the expression of ‘X’ is realized as its oblique object, introduced by the preposition ʔə

(21c).

2.1.1.7 Desiderative -ab

The morphological desiderative is formed with the suffix -ab ‘desiderative [DSD]’, which is

used in conjunction with the prefix dxʷ(s)- (Section 2.1.1.1); together, these affixes take a verb

stem meaning ‘X’, as in (22a), and form a desiderative stem meaning ‘want to X’, (22b):

(22) a. ʔu siʔiʔab tux%ʷ čələp ɬuhəliʔdxʷəxʷ kʷi dbədbədaʔ

ʔu siʔ–iʔab tux%ʷ čələp ɬu=həliʔ–dxʷ=əxʷ kʷi d–bəd–bədaʔ

INTJ PL–noble PTCL 2PL.SUB IRR=alive–DC=now REM 1SG.PO–DSTR–offspring

‘oh, sirs, you will save my children [from starvation]!’

(Hess 1998: 80, line 70)

b. yəx%i čəd huy ʔəxʷəliʔdubutəb

yəx%i čəd huy ʔəs–dxʷ–həliʔ–dxʷ–but–əb

because 1SG.SUB SCONJ STAT–CTD–alive–DC–REFL–DSD

‘because I want to cure myself [from a sickness]’

(Hess 1998: 58, line 50)

(22a) shows the verb həliʔdxʷ ‘cure someone, save someone’, while (22b) shows the same verb

in its desiderative form, dxʷhəliʔdxʷ ‘want to cure someone, want to save someone’. The verb in

(22b) is also in the reflexive, illustrating the position of the desiderative affix at the end of the

suffix string. This is also seen in (23):

(23) ʔudxʷslək’ʷdxʷyitəbab dᶻəɬ tsi x%ənimulicaʔ ʔə kʷi sɬuʔuməs, təluməs

ʔu–dxʷs–lək’ʷ–dxʷ–yi–t–əb–ab dᶻəɬ tsi x%ənimulicaʔ

PFV–CTD–eaten–DC–DAT–ICS–PASS–DSD PTCL SPEC:FEM x%ənimulicaʔ

ʔə kʷi sɬuʔb–s t’əluʔb–s

PR REM chum–3PO dried.king.salmon–3PO

‘they want to eat x&ənimulicaʔ’s chum and dried king salmon’

(Hess 1998: 57, line 40)

The prefixal forms dxʷ- and dxʷs- seem for the most part to be in free variation, although according to Hess

(1998) there are a few texts in which one or the other of the two forms was deemed unacceptable with a particular

root in a particular context. More research is needed to account for these discrepancies.

These last two words are Raven’s rendering of sɬuʔbs ‘his/her chum’ and t’əluʔbs ‘his/her dried king salmon’.

x&ənimulicaʔ is the name sometimes given to Crow in traditional stories.

Here, -ab follows the passive marker, -b, occupying ultimate final position. Note, however, that

the prefix dxʷ(s)- continues to occupy its normal position directly adjacent to the stem, inside the

aspectual prefixes.

The desiderative affix has two phonological allomorphs, [-ab] and [-əb]. The former appears

bearing stress in stems that have only schwa, while the latter appears in unstressed position

following stems that contain non-schwa vowels:

(24) a. ʔəxʷsʔəɬədáb čəd

ʔəs–dxʷs–ʔəɬəd–ab čəd

STAT–CTD–eat–DSD 1SG.SUB

‘I want/need to eat’

b. ʔəxʷsʔitutəb čəd

ʔəs–dxʷs–ʔitut–b čəd

STAT–CTD–NM–sleep–DSD 1SG.SUB

‘I want/need to sleep’

(Hess 1995: 47)

As shown by the glosses of these examples, desiderative stems can have either the reading ‘want

to X’ or ‘need to X’, context providing the necessary disambiguation.

One particularly frequent use of the desiderative suffix is in the lexicalized expression

dxʷcutəb ‘think something (lit. ‘want to say something’):

(25) ʔəxʷcutəb čəd ʔə kʷi dsəsx%əɬ

ʔəs–dxʷ–cut–b čəd ʔə kʷi d=s=ʔəs–x%əɬ

STAT–CTD–speak–DSD 1SG.SUB PR REM 1SG.PO=NM=STAT–sick

‘I think that I am sick’

(Bates, Hess & Hilbert 1994: 48)

This is the only expression in which the addition of the desiderative seems to have any effect on

the valency of its base, changing the monovalent intransitive radical √cut ‘speak’ to a bivalent

intransitive dxʷcutəb ‘think something’; it is also possible that, etymologically, the stem was

formed on the transitivize cut-t, formed with the internal causative, this affix being “absorbed”

into the stem by degemination. The fact that this is synchronically a fossilized form is

highlighted by the existence of a derived form dxʷcutəbid ‘realize something’ in which the

derivational affixal complex -bi-d (Section 2.1.3.2) follows the desiderative suffix,

demonstrating that it has become a part of the verb stem. Note that the fusion of the stative

aspectual prefix and the prefixal portion of the desiderative marker is the same morphophonemic

process undergone by dxʷ(s)- in all other environments (see Section 2.1.1.1 above).

2.1.1.8 Partitive ʔiɬ-

The prefix ʔiɬ- ‘partitive [PRTV]’ is affixed primary (but not exclusively) to verbal bases to

convey the notion that only some of a possible number of event-participants or semantic actants

are involved in the event expressed by the clause in which it occurs. In the prefix string, the

partitive appears preceding any aspectual prefixes and following proclitics such as the habitual

ƛ’u= and the nominalizer s= in the following example:

(26) diɬ šəbab ʔə tiʔəʔ qʷiqʷistay’bixʷ ƛ’usʔiɬux%ilix%s haʔkʷ tiʔiʔiɬ

diɬ šəbab ʔə tiʔəʔ qʷiqʷ–qʷistay’bixʷ ƛ’u=s=ʔiɬ–ʔu–x%ilix%=s

FOC enemy PR PROX DSTR–dwarf HAB=NM=PRTV–PFV–battle=3PO

haʔkʷ tiʔiʔiɬ

long.time DIST:PL

‘it was they who were the enemies of the dwarves on some of whom they had long been

making war’

(Hess 2006: 65, line 556)

When the prefix occurs with intransitive verbs, it expresses the fact that the subject of the verb

expresses only a subset of a group of potential ACTORs:

(27) a. x%ʷul’ čəxʷ ʔiɬʔəɬəd

x%ʷul’ čəxʷ ʔiɬ–ʔəɬəd

only 2SG.SUB PRTV–eat

‘you just [go ahead and] eat [without me/us]’

b. ʔəca kʷi ɬuʔiɬsula

ʔəca kʷi ɬu=ʔiɬ–sula

I REM IRR=PRTV–be.toward.centre

‘the one who will be towards the centre is me’

(Hess 1998: 35)

With transitive verbs, ʔiɬ- expresses the fact that the action affects only a subset of possible

UNDERGOERs:

(28) a. x%ʷul’ ʔiɬčəbaʔəd

x%ʷul’ ʔiɬ–čəbaʔ–d

only PRTV–laden–ICS

‘he carried just some of it on his back’

(Hess 1998: 35)

b. x%ʷul’ ʔiɬdᶻak’ʷadiʔəd tiʔəʔ suq’ʷsuq’ʷas, tiʔəʔ ʔalalš

x%ʷul’ ʔiɬ–dᶻak’ʷadiʔ–d tiʔəʔ suq’ʷ–suq’ʷas–s tiʔəʔ ʔal–alš–s

only PRTV–invite–ICS PROX DSTR–younger.brother PROX PL–cross.sex.sibling–3PO

‘he invited just his younger brothers, his siblings’

[DM Basket Ogress, line 4]

The partitive has the same meaning when affixed directly to nominal roots when these are

sentence predicates:

(29) a. x%ʷul’əxʷ ʔiɬp’uay’, ʔiɬtulqʷ, ʔiɬbəsqʷ

x%ʷul’=əxʷ ʔiɬ–p’uay’ ʔiɬ–tulqʷ ʔiɬ–bəsqʷ

only=now PRTV–flounder PRTV–mussel PRTV–crab

‘it was only some flounder, some mussels, some crab’

(Hess 1998: 35)

b. ʔuqʷuʔqʷadid kʷi sʔəx%ʷa ʔə kʷi sʔiɬləgʷəbs

ʔu–qʷuʔqʷa–di–d kʷi sʔəx%ʷa ʔə kʷi s=ʔiɬ–ləgʷəb=s

PFV–drink–SS–ICS REM urine PR REM NM=PRTV–youth=3PO

‘he has drunk the urine of the one who is his fellow-youth’

(Hess 1998: 68, line 97)

c. x%ʷul’ xʷəlušəd tiʔəʔ dsuʔabyitəb ʔə tiʔəʔ dsʔiɬwiw’su

x%ʷul’ xʷəlušəd tiʔəʔ d=s=ʔu–ʔab–yi–t–əb ʔə tiʔəʔ

only fish.tail PROX 1SG.PO=NM=PFV–extend–DAT–ICS–PASS PR PROX

d=s=ʔiɬ–wiw’su

1SG.PO=NM=PRTV–children

‘what I am given by the other children is just fish tail’

(lit. ‘what I am given by these who are my fellow-children is just fish-tail’)

[AJ Basket Ogress, line 33]

The sentence in (29a) consists of a series of clauses with nominal predicates and zero subjects.

The effect of the partitive in (27a) is, as in the expressions in (26), to show that the subject

corresponds to only a sub-part of the group of potential ACTORs/subjects — of the flounder, only

a part of it, of the mussels, only a portion, etc. In the next two examples, ʔiɬ- appears on a noun

serving as the predicate of an embedded clause. The agentive complement (the AGENT) of the

passivized verb ʔabyitəb ‘be given something’ in (29c) is tiʔəʔ dsʔiɬwiw’su ‘those who are my

fellow-children’, a nominalization of dʔiɬwiw’su Ø ‘they [are] my fellow children’. In (29c), the

partitive thus singles out only a sub-group of the children — specifically, those that are not ‘I’.

Another common usage of the partitive is in modificative constructions such as those in (30):

(30) huy, ƛ’uq’ʷuʔcutəxʷ tiʔəʔ ʔiɬluƛ’luƛ’ wiw’su, ʔiɬqʷiq’ʷqʷiq’ʷ wiw’su

huy ƛ’u=q’ʷuʔ–t–sut=əxʷ tiʔəʔ ʔiɬ–luƛ’–luƛ’ wiw’su

SCONJ HAB=gather–ICS–REFL=now PROX PRTV–DSTR–old children

ʔiɬ–qʷiq’ʷ–qʷiq’ʷ wiw’su

PRTV–DSTR–strong children

‘so then the older children, the stronger children got together’

[AJ Basket Ogress, line 93]

This example shows two monovalent intransitive radicals that express property concepts acting

as (pre-posed) relative clauses, the partitive prefix in these cases indicating the sub-division of

each group to whom the property pertains — ʔiɬluƛ’luƛ’ wiw’su ‘the older children’ (or, more

precisely, ‘the children that are older than the others’), and ʔiɬqʷiq’ʷqʷiq’ʷ wiw’su ‘the stronger

children’ (‘the children that are stronger than the others’). Because of the subdivision that the

partitive makes within the class of things referred to by the head noun, expressions such as these

entail an implicit comparison for the property expressed by the modifying predicate. This

implication of ʔiɬ- manifests itself in the appearance of the partitive in comparative

constructions:

(31) ʔiɬxʷaʔxʷaʔ tiʔəʔ dsdəxʷiɬ dxʷʔal adsgʷaʔ

ʔiɬ–xʷaʔxʷaʔ tiʔəʔ d–sdəxʷiɬ dxʷ–ʔal ad–sgʷaʔ

PRTV–light.weight PROX 1SG.PO–hunting.canoe CNTRPT–at 2SG.PO–one’s.own

‘my hunting canoe is lighter than yours’

(Hess 1998: 36)

Comparatives are discussed in more detail in Section 8.8 below.

David Beck 10-2-7 2:19 PM

Comment: nominalization is required here

because it is a predicate nominal???

Another related use of the partitive involves its affixation to gradable predicates to express a

comparatively higher degree, as in (32):

(32) xʷul’ ʔiɬdukʷtub

xʷul’ ʔiɬ–dukʷ–txʷ–b

only PRTV–anormal–ECS–PASS

‘the just got madder at them’

(Hess 1998: 83, line 156)

The same use of the partitive is found even more commonly in locative expressions to indicate

location farther along in a particular direction or motion part way to a destination:

(33) a. ʔa kʷi ƛ’ubəsʔiɬt’aq’ts, liɬʔal tiʔiɬ ʔiɬt’aq’t

ʔa kʷi ƛ’u=bə=s=ʔiɬ–t’aq’t=s liɬʔal tiʔiɬ ʔiɬ–t’aq’t

be.there REM HAB=ADD=NM=PRTV–landward=3PO PR DIST PRTV–landward

‘there he would be again on the landward [side], towards the landward [side]’

(Hess 1998: 65, line 16)

b. x%ʷul’əxʷ ʔiɬʔux%ʷ tsiʔəʔ diʔəʔ sk’ʷuys dxʷʔal tiʔəʔ haʔɬ šəgʷɬ

x%ʷul’=əxʷ ʔiɬ–ʔux%ʷ tsiʔəʔ diʔəʔ sk’ʷuy–s dxʷ–ʔal tiʔəʔ haʔɬ šəgʷɬ

only=now PRTV–go DIST:FEM here mother–3PO CNTRPT–at PROX good path

‘his mother just went up the good path’

[AW Basket Ogress, line 39]

This use of the partitive is largely restricted to verbal radicals, although there is one such

expression in the corpus based on a noun, ʔilaq ‘stern of canoe’:

(34) gʷəl x%ʷul’əxʷ ləƛ’čabcut tiʔəʔ cədiɬ ʔiɬʔilaq

gʷəl x%ʷul’=əxʷ lə=ƛ’čab–t–sut tiʔəʔ cədiɬ ʔiɬ–ʔilaq

SCONJ only=now PROG=double.over–ICS–REFL PROX he PRTV–stern

‘and this one in the stern’

(Hess 2006: 52, line 234)

At this point it is not clear if this is a lexicalized expression, a more general use of the partitive

with particular types of partonymic expressions (or specifically, with vocabulary for parts of a

canoe), or if it represents a productive, if infrequent, use of ʔiɬ- with locative nouns.

ʔiɬ- is also frequently used with nouns referring to kinship or age-groups of humans and, in at

least one case, types of animal from the time of legends:

(35) a. gʷəl tsiʔəʔ ʔiɬt’isuʔ gʷəl ƛ’uləqdxʷ tsiʔiɬ luƛ’

gʷəl tsiʔəʔ ʔiɬ–t’isuʔ gʷəl ƛ’u=ləq–dxʷ tsiʔiɬ luƛ’

SCONJ PROX:FEM PRTV–younger.relative SCONJ HAB=hear–DC DIST:FEM old

‘as for the younger of the relatives, she overheard the old woman’

[DS Star Child, line 122]

b. diɬ tugʷəɬ tiʔiɬ ʔiɬsqigʷac ʔi tiʔəʔ cədiɬ bəščəb tiʔəʔ tuqəliltub

diɬ tu=gʷəɬ tiʔiɬ ʔiɬ–sqigʷac ʔi tiʔəʔ cədiɬ bəščəb

FOC PAST=belong.to DIST PRTV–deer and PROX he mink

tiʔəʔ tu=qəl–il–txʷ–b

PROX PAST=bad–INCH–ECS–PASS

‘they belonged to Deer and Mink who had been warned to stop’

[ML Mink and Tutyika I, line 230]

The effect of ʔiɬ- in such expressions is to subdivide a group of humans (or, in the case of 35b,

animals) according to the meaning of the nominal base. Thus, in (35a), the partitive expresses

that of the (two blood-related) women present, the one who overhead the old woman was the

younger. Likewise, the partitive expression ʔiɬsqigʷac in (35b) singles out, from amongst the

animals in the story, the one who is a deer.

Such forms can be used nominally, as in (35), or as

parts of nominalized clausal expressions, as in (29b) and (c) above.

Three particularly common partitive expressions are ʔiɬdᶻixʷ ‘first; better, best’, ʔiɬlaq ‘later,

last’, and ʔiɬkʷəlq ‘others’. The first of these three is based on an adverb, dᶻixʷ ‘first’ and, when

combined with the partitive, has two senses— a literal ‘first of all’ and a more figurative ‘better,

best’:

(36) a. gʷəl dəgʷi kʷi ɬukʷədatəb dᶻixʷ

gʷəl dəgʷi kʷi ɬu=kʷəda–t–əb dᶻixʷ

then you REM IRR=taken–ICS–PASS first

‘well then the one that will be taken first is you’

[LA Basket Ogress, line 26]

b. tiˑləb kikəwič tiʔiɬ ʔiɬdᶻixʷ

tiˑləb ki–kəwič tiʔiɬ ʔiɬ–dᶻixʷ

immediately ATTN–hunchback DIST PRTV–first

‘right then Little Hunchback was the first [taken]’

[LA Basket Ogress, line 101]

Or, more precisely, the one who became a deer. This line is from a story involving dukʷibəl, the Changer, who

created animals from the first people depending on what they were found doing at the time he passed by.

c. ʔiɬdᶻixʷ čəxʷ ʔəslax%dxʷ

ʔiɬ–dᶻixʷ čəxʷ ʔəs–lax%–dxʷ

PRTV–first 2SG.SUB STAT–remember–DC

‘you remember better’

(Bates, Hess & Hilbert 1994: 91)

As shown by (36c) in particular, the base’s part of speech is unaffected by the addition of the

partitive prefix: ʔiɬdᶻixʷ, like ʔiɬdᶻixʷ, is an adverb.

The polar opposite expression, ʔiɬlaq ‘later, last’, on the other hand, appears to be based on a

verbal radical, √laq ‘be last, be behind’:

(37) a. huy ʔaləxʷ tiʔiɬ laqəxʷ

huy ʔal=əxʷ tiʔiɬ laq=əxʷ

SCONJ PR=now DIST last=now

‘then [they came] to the last one [obstacle]’

(Hess 2006: 36, line 357)

b. diɬəxʷ yuwaɬ ʔiɬlaq

diɬəxʷ yuwaɬ ʔiɬ–laq

FOC=now ultimate PRTV–last

‘she was the very last one’

(Hess 1998: 73, line 209)

c. ɬusuʔəɬəd čəxʷ ʔə kʷi ʔiɬlaq ʔaciɬtalbixʷ

ɬu=s=ʔu–ʔəɬəd čəxʷ ʔə kʷi ʔiɬ–laq ʔaciɬtalbixʷ

IRR=NM=PFV–eat 2SG.SUB PR REM PRTV–last person

‘you are what later people will eat’

(Hess 1998: 75, line 251)

(37a) shows √laq on its own as the head of a referential expression (that is, a headless relative

clause).

In (37b) it acts as a sentence predicate, while in (37c) it appears as a modifier, part of

an idiomatic expression, ʔiɬlaq ʔaciɬtalbixʷ ‘the last people’, referring to the modern people of

today’s world (as opposed to the people of the time of legends). Once again, affixation with ʔiɬ-

appears to have no effect on the part of speech of its base.

The last of these three expressions, ʔiɬkʷəlq ‘others’, is a little harder to pin down as it is

unattested on its own in contextualized form in the present corpus:

This is, of course, also a possible position for an adverb, raising the possibility that √laq does belong to the same

part of speech as dᶻixʷ. However, unlike dᶻixʷ, √laq is not attested in pre-verbal position, and it is found combined

with verbal derivational morphemes (e.g., laqil ‘be late’) that are not generally associated with adverbs.

(38) a. kʷəlq

kʷəlq

others

‘other things’

(Bates, Hess & Hilbert 1994: 125)

b. tubəy’dubəxʷ tiʔəʔ cədiɬ ʔiɬkʷəlq

tu=bə=ʔəy’dxʷ–b=əxʷ tiʔəʔ cədiɬ ʔiɬ–kʷəlq

PAST=ADD=find–PASS=now PROX he PRTV–other

‘he found some others’

(Hess 1998: 75, line 257)

It is difficult to surmise from examples such as (38a) what part of speech the radical kʷəlq

belongs to. Judging from the distribution of the partitive form, ʔiɬkʷəlq, it is possible that it is

either verbal or nominal; however, given that all of the textual attestations of ʔiɬkʷəlq are in

referential expressions such as that in (38b) (and the absence of attestations of ʔiɬkʷəlq as a

sentence predicate), it may be that ʔiɬkʷəlq — and, by extension, kʷəlq — is best treated as a

noun. It should also be noted that kʷəlq has a plural, kʷaalq ‘others [people]’, formed by -V

plural reduplication.

2.1.2 Valency-increasing verbal affixes

Because the bulk of Lushootseed radicals are monovalent and intransitive, the language

requires a rather extensive inventory of derivational means to form verb stems expressing events

with multiple participants or semantic actants. In total there are six affixes that are used

exclusively to increase the valency of an intransitive stem, as well as a four secondary suffixes

that are used in conjunction with one of these affixes. The primary distinction that can be drawn

among the valency-increasers is that between causatives and applicatives. A causative affix is

one that adds a new event-participant (semantic actant) which is expressed as a syntactic subject.

In most languages, the semantic role of this actant is that of CAUSER; however, Lushootseed does

not distinguish the role of CAUSER from that of AGENT, and so many (if not most) of the subjects

added by Lushootseed causatives express AGENTs. Applicatives, on the other hand, add a new

semantic actant which is expressed by an object and which is assigned a variety of semantic

roles. As in most languages, the semantic roles assigned to applicative objects in Lushootseed are

rather diverse, but in most cases Lushootseed applicatives do not assign the role of PATIENT.

Within the causatives, it is possible to distinguish affixes according to the government

patterns of the stems they create — that is, according to the different grammatical relations

assigned to their syntactic arguments. This leads to a distinction between transitive causatives,

causatives that create transitive stems that take a direct object, and intransitive causatives,

causatives that create bivalent intransitive verb stems that take an oblique object. Applicatives

can in principle be sub-categorized according to this criterion as well, although all applicative

affixes in Lushootseed fall into the category of transitive applicatives in that they subcategorize

for direct objects. Both types of affix can then be further subdivided according to additional

semantic criteria. Lushootseed has, for example, three transitive causative affixes, -t ‘internal

causative’, -txʷ ‘external causative’, and -dxʷ ‘diminished control causative’, all of which have

the same syntactic effect on the stem but which express events with different semantic

characteristics. A list of valency-increasing affixes categorized according to the taxonomy

proposed here is given in Table 19:

affix

name

affix-type

type of stem

-t

‘internal causative’

transitive causative

transitive

-txʷ

‘external causative’

transitive causative

transitive

-dxʷ

‘diminished control causative’

transitive causative

transitive

-b

‘causative middle’

intransitive causative

bivalent intransitive

-alikʷ

‘causative of activity’

intransitive causative

bivalent intransitive

-c/-s

‘allative applicative’

transitive applicative

transitive

-yi-

‘dative applicative’

transitive applicative

trivalent transitive

-bi-

‘middle applicative’

transitive applicative

transitive

-di-

‘secondary suffix’

transitive applicative

transitive

-i-

‘secondary suffix’

transitive applicative

transitive

Table 19: Valency-increasing affixes

Of the affixes listed here, the first eight are well-attested as analyzable parts of a substantial

number of lexemes; the last two secondary suffixes are less productive, although they form part

of a few high-frequency lexical items. Each of these affixes will be discussed in turn in the

following sections.

2.1.2.1 Internal causative -t

By far the most prevalent of the valency-increasing verbal affixes is -t ‘internal causative

[ICS]’. This suffix is not only the most frequent in terms of the number of stems of which it forms

a part, but it is also the most complex in terms of its allomorphy and morphophonemics and the

most varied in terms of its syntactic effects on the stem to which it attaches. Its primary and most

prevalent use is as a transitive causative suffix which changes a patient-oriented monovalent

stem expressing a state into a transitive stem by adding a semantic AGENT, realized as a syntactic

subject:

(39) a. ʔut’uc’ čəd

ʔu–t’uc’ čəd

PFV–shot 1SG.SUB

‘I got shot’

(Bates, Hess & Hilbert 1994)

b. ʔut’uc’ucid ʔu

ʔu–t’uc’u–t–sid ʔu

PFV–shot–ICS–2SG.OBJ INT

‘did s/he shoot (at) you?’

(Hess 1995: 43, ex. 11a)

c. ʔut’uc’utəb čəd

ʔu–t’uc’u–t–əb čəd

PFV–shot–ICS–PASS 1SG.SUB

‘I was shot (at)’

(Bates, Hess & Hilbert 1994)

As shown in these examples, a monovalent patient-oriented radical such as t’uc’ ‘be shot; be the

target of a missile’ takes the internal causative suffix to form a transitive verb, t’uc’ud ‘shoot

someone; shoot at someone’ (or, more literally, ‘cause to be shot; cause to be the target of a

missile’). While the vast majority of stems formed with -t have translation equivalents in most

languages which are ordinary transitive verbs, the syntactic effects of this suffix are clearly

causative, as are its semantic effects — keeping in mind the basically stative nature of

Lushootseed radicals: the radical itself expresses a state while the derived transitive stems

expresses an action preformed by an AGENT resulting in a PATIENT coming into that state.

This

added AGENT, like the CAUSER in typologically more ordinary causatives, is realized as the

syntactic subject while the subject of the radical becomes the direct object of the transitive stem,

marked by the s-series of object markers (Section 8.1.2), as in (39b). As shown by (39c), the

object of an internal causative stem is an ordinary direct object and is amenable to syntactic

operations such as passivization. When the derived stem takes an overt NP argument, this

argument is obligatorily interpreted as direct object:

(40) ʔuk’ʷəɬəd ti qʷuʔ

ʔu–k’ʷəɬ–d ti qʷuʔ

PFV–poured–ICS SPEC water

‘s/he poured the water’

(Hess 1995: 18, ex. 1a)

This is an interpretative property of all transitive verbs and will be discussed in more detail in

Section 8.2.2 below.

The internal causative suffix -t has four allomorphs — [-t], [-d], [-əd], and [-š]. Of these, [-t],

[-d], [-əd] are phonologically conditioned: [-t] is the elsewhere form, while [-d] is restricted to

word-final position (that is, last position suffix-string not including bound enclitics) following

vowels and approximants (41):

(41) qʷuʔqʷad ‘drink something’ < √qʷuʔqʷa ‘have a drink’

ɬild ‘give food to’ < √ɬil ‘make a gift of food’

qʷ’əld ‘cook something’ < √qʷ’əl ‘be cooked, be ripe’

For further discussion, see Beck (1996). The distinction between the internal causative -t and the external

causative -txʷ will be taken up again in the next section (2.1.2.2).

Note that there are very few vowel-final radicals in Lushootseed and of these few three are exceptional. Two of

them, √lu ‘be heard’ and √gʷi ‘make an invitation’, undergo vowel-lengthening when the internal causative is added

to give the forms luud ‘hear something’ and gʷiid ‘call someone’. The remaining radical, √ʔa ‘be there’ has the

transitive form ʔaʔəd ‘put something in a place’, formed by inserting a glottal stop into the coda of the radical and

applying the [-əd] allomorph of the internal causative. As noted in Section 2.1.1.3, for some speakers the radical is

√ʔaʔ ‘be there’ instead of √ʔa, so ʔaʔəd diachronically may have been a regular form.

šuk’ʷild ‘grey someone’ < šuk’ʷil ‘turn grey’

ʔaʔild ‘put away’ < ʔaʔil ‘get caught’

xʷit’ild ‘lower something’ < xʷit’il ‘climb down’

h$dʔiw’d ‘bring inside’ < √h$dʔiw’ ‘be inside a house’

[-əd] is also restricted to word-final position and appears following obstruents, as in (42):

(42) č$baʔ$d ‘backpack something’ < √č$baʔ ‘be loaded down with something’

ƛ’iq’alusəd ‘stick eyes shut’ < √ƛ’iq’ ‘be stuck, be sticky’ + -alus ‘eye’

pədičəd ‘dirty something’ < √pəd ‘be dirty, be buried’ + -ič ‘covering’

qiq’əd ‘confine someone’ < √qiq’ ‘be confined’

šəqəd ‘move up high’ < √šq ‘be high’

qʷšabəd ‘fog something up’ < √qʷšab ‘be foggy’

həliʔəd ‘cure someone’ < √həliʔ ‘be alive’

xʷəcəd ‘remove something’ < √xʷəc ‘be removed’

bəlx%ʷəd ‘pass someone’ < √bəlx%ʷ ‘be beyond’

The [-əd] allomorph is replaced by [-t] (rather than [-ət]) in non-final position, as shown by the

examples in (43):

(43) a. ʔabcut (cf. ʔabəd ‘extend something’)

ʔab–t–sut

extend–ICS–REFL

‘it extends itself’

(Hess 2006: 50, line 216)

b. ʔabaqcid (cf. ʔabaqəd ‘return something’)

ʔabaq–t–sid

return–ICS–2SG.OBJ

‘s/he returns you’

(Hess 2006: 29, line 170)

c. x%əctəb (cf. xʷəcəd ‘remove something’)

x%əc–t–əb

removed–ICS–PASS

‘it was removed’

(Hess 2006: 79, line 896)

d. ƛ’iq’ƛ’iq’alustəb (cf. ƛ’iq’alusəd ‘stick someone’s eyes shut’)

ƛ’iq’–ƛ’iq’–alus–t–əb

DSTR–sticky–eye–ICS–PASS

‘their eyes were stuck shut’

[DM Basket Ogress, line 13]

However, šuk’ʷiləd ‘grey someone’ is also attested.

An exception to this generalization is kʷaʔd ‘let go of’, which is only infrequently attested as kʷ aʔəd; note also

the verb cut ‘speak to someone’ from the radical √cut ‘speak’, which does not have the expected form *cutəd.

David Beck 10-2-7 2:19 PM

Comment: what about after the fake h in

?ahed?

As seen in (43d), the allomorphy of the -t form of the internal causative is the same both when it

is affixed directly to a radical and when it follows another type of affix such as a lexical suffix.

It should be noted here that the schwa associated with the [-əd] allomorph of internal

causative differs from the epenthetic vowels associated with the internal causative forms of

certain types of radicals discussed at the beginning of Section 2. Unlike the schwa in [-əd], true

epenthetic vowels are maintained in the presence of subsequent affixes, as shown by the passive

forms in (44):

(44) a. ʔukʷədad ti sqʷəbayʔ

ʔu–kʷəda–d ti sqʷəbayʔ

PFV–held–ICS SPEC dog

‘s/he grabbed the dog’

b. ʔukʷədatəb ʔə tsi č’ač’as ti sqʷəbayʔ

ʔu–kʷəda–t–b ʔə tsi č’ač’as ti sqʷəbayʔ

PFV–held–ICS–PASS PR SPEC:FEM child SPEC dog

‘the dog was grabbed by the girl’

(Hess 1995: 22, ex. 5b – c)

c. xʷəbəd

xʷəb–əd

thrown–ICS

‘he tossed it’

(Hess 2006: 55, line 327)

d. xʷəbtəb dxʷʔal ti x%ʷəlč

xʷəb–t–b dxʷ–ʔal ti x%ʷəlč

thrown–ICS–PASS CNTRPT–at SPEC sea

‘[she] was thrown into the sea’

(Hess 1998: 75, line 252)

The stem in (44a), kʷədad ‘take something’ is formed from the radical √kʷəd ‘be taken’, a C$C

radical that takes an epenthetic /a/ vowel in its internal causative form; this epenthetic /a/ persists

even in the passive, kʷədatəb ‘be taken’ (44b). The epenthetic schwa found with xʷəbəd ‘toss

something’ (based on √xʷəb ‘be thrown’), on the other hand, disappears in the passive xʷəbtəb

‘be thrown’. The most straightforward treatment of this phenomenon is to consider the schwa in

xʷəbəd to be part of a word-final allomorph of the internal causative marker, and to contrast it

with the epenthetic vowel found with the internal causative forms of radicals like √kʷəd. The

latter can thus be considered a part of a morphologically-conditioned allomorph [kʷəda] of the

radical itself associated with the internal causative (and causative middle) morpheme.

The third allomorph of the internal causative, the suppletive form [-š], is restricted to a

handful of stems which either require [-š] or have two attested forms, one with [-š] and the other

with [-t]. Those stems that are only attested with [-š] are given in Table 20:

ʔišɬš ‘paddle  [canoe]’

(√ʔišɬ ‘paddle canoe’)

ɬalš ‘remove  from fire’

(º√ɬal ‘be removed from fire’; cf. ɬaldxʷ ‘manage to get  out of fire’)

ƛ’alš ‘wear ’

(√ƛ’al ‘put  on’)

suxʷtəš ‘recognize ’

(º√suxʷt ‘know identity’; cf. suxʷtil ‘recognize’)

tagʷš ‘buy ’

(√takʷ ‘be bought’)

təbaš ‘crave ’

(*√təb ‘have a craving’)

Table 20: Stems requiring the -š allomorph of the internal causative

Four of the six stems in this set depart from the basic internal-causative pattern illustrated in

(39), whereby an intransitive radical is causativized by adding a semantic AGENT expressed as a

syntactic subject. The verb ʔišɬš ‘paddle something [canoe]’ follows the applicative pattern

shown by the set of stems given in Table 20, while the verbs ƛ’alš ‘wear something’, suxʷtəš

‘recognize someone’, and təbaš ‘crave something’ are merely transitivized forms of bivalent

intransitive radicals. The stems in Table 20 also show idiosyncratic behaviour in their passive

forms (Section 6.2). suxʷtəš ‘recognize someone’ forms its passive with [-t] instead of [-š], while

the passive of ɬalš ‘remove something from fire’ is attested as both ɬalšəb and ɬaltəb. Two of

these stems, ƛ’alš ‘wear something’ and tagʷš ‘buy something’, have idiosyncratic passive forms

in -ib (ƛ’alib ‘be worn’ and tagʷib ‘be bought’, respectively). ʔišɬš ‘paddle something [canoe]’

has no passive formed on the internal causative, speakers opting instead for a passive form based

on the external causative -txʷ, ʔišɬtub ‘be paddled’. The verb təbaš ‘crave something’ has no

attested passive form in the present corpus. Two of the stems in Table 20 — ɬalš ‘remove

something from fire’ and təbaš ‘crave something’ — have no independently attested radicals and

so would have to be treated synchronically as inherently transitive stems. They are included here

for the sake of completeness.

Another slightly larger set of stems has both [-š] and [-t] forms, choice between the two

depending to some extent on dialect (the [-š] form is more frequent in the Skagit dialect than in

Snohomish) and on the individual speaker. These stems are given in Table 21:

ʔabš ‘extend , give ’

ʔabəd ‘extend , give ’

ʔabš ‘give ’

ʔabad ‘give ’

bəčaš ‘set  down’

bəčad ‘set  down’

c’agʷš ‘wash ’

c’agʷad ‘wash ’

dəgʷaš ‘put  inside’

dəgʷad ‘put  inside’

ƛ’agʷš ‘stitch  (mat)’

ƛ’agʷəd ‘stitch  (mat)’)

ɬaq’aš ‘put  down’

ɬaq’ad ‘put  down’

p’t’aš store , tidy ’

p’t’ad ‘store , tidy ’

qʷataš ‘lay  out’

qʷatad ‘lay  out’

t’agʷtəš ‘put  on top’

t’agʷtəd ‘put  on top’

x&alš ‘write ’

x&alad ‘write ’

Table 21: Internal causative stems with -š and -t forms

Like [-t], the [-š] allomorph requires an epenthetic schwa following certain obstruents —

specifically, voiceless stops in word-final position (e.g., t’agʷtəš ‘put something on top’ from

t’agʷt ‘be on top’);

this contrasts somewhat with the morphophonemic behaviour of [-t], which

requires epenthesis after all obstruents in word final position, as in (42) above (cf. c’agʷš ‘wash

something’ vs. c’agʷad ‘wash something’, both based on √c’aʔkʷ ‘be washed’). The passives of

these verbs are all based on the [-t] form of the stem.

In addition to its morphophonemic interactions with stems and radicals, the internal causative

shows a certain amount of variability in terms of its syntactic effects on the valency and

government pattern of its base. In the overwhelming majority of cases, the internal causative

follows the pattern illustrated in (39) — that of a causative morpheme that adds a semantic

AGENT/syntactic subject, augmenting the valency of the verb by one. The forms given in Table

This is also seen in suxʷtəš ‘recognize someone’ in Table 20.

22 are a representative sample of internal causative stems based on radicals attested as free

forms:

ʔaʔəd ‘put  there’

(√ʔa ‘be there’)

ʔix&ʷid ‘throw  away’

(√ʔix&ʷ ‘be thrown; have thrown to’)

bapad ‘pester ’

(√bap ‘be busy’)

bəčad ‘set  down’

(√bəč ‘be lying, be fallen from standing’)

caq’ad ‘spear ’

(√caq’ ‘be speared’)

cilid ‘dish  out’

(√cil ‘be dished up’)

c’agʷad ‘wash ’

(√c’aʔkʷ ‘be washed’)

čalad ‘chase ’

(√čal

‘be overtaken’)

č’axʷad ‘hit with a stick’

(√č’axʷ ‘be hit with a stick’)

daʔad ‘name ’

(√daʔ ‘be named’)

dəgʷad ‘put  inside’

(√dəkʷ ‘be inside’)

dᶻakʷad ‘rock ’

(√dᶻakʷ ‘be shaky, be shaking’)

dᶻaƛ’əd ‘confuse ’

(√dᶻaƛ’ ‘be confused’)

dᶻalqəd ‘turn  around’

(√dᶻal ‘turn around, turn over’)

dᶻix&id ‘break  down, take  apart’

(√dᶻix& ‘be broken down, be fallen apart’)

gʷəx&ad ‘untie ’

(√gʷəx& ‘be untied’)

huyud ‘make ’

(√huy ‘be done, be made, be finished’)

kʷədad ‘take ’

(√kʷəd ‘be held, be taken’)

lək’ʷəd ‘eat ’

(√lək’ʷ ‘be eaten’)

lild ‘move away’

(√lil ‘be far’)

ɬaq’ad ‘put  down’

(√ɬaq’ ‘be fallen, be lying down’)

ɬič’id ‘slice ’

(√ɬič’ ‘get cut with knife’)

ɬidid ‘tie ’

(√ɬid ‘be tied’)

ƛ’iqid ‘take  out from within’

(√ƛ’iq ‘emerge’)

pədičəd ‘dirty ’

(from √pəd ‘be dirty, be buried’ + -ič ‘covering’)

pusud ‘throw at ’

(√pus ‘be hit by  (missile)’)

qiq’əd ‘confine ’

(√qiq’ ‘be confined’)

qʷatad ‘lay  out’

(√qʷat ‘be lying; snow falls’)

qʷibid ‘prepare ’

(√qʷib ‘be ready’)

qʷšabəd ‘fog  up’

(√qʷšab ‘be foggy’)

q’axʷad ‘freeze ’

(√q’axʷ ‘be frozen’)

q’ilid ‘put  on board’

(√q’il ‘be aboard’)

q’ʷəld ‘cook ’

(√q’ʷəl ‘be cooked, be ripe’)

q’ʷibid ‘unload  from conveyance’

(√q’ʷib ‘be disembarked, be unloaded’)

q’ʷuʔəd ‘gather ’

(√q’ʷuʔ ‘be together with ’)

šəqəd ‘move  up high’

(√šq ‘be high’)

šubud ‘make  disappear; massacre ’

(√šub ‘disappear’)

šuk’ʷild ‘grey ’

(šuk’ʷil ‘turn grey’ from √šuk’ʷ ‘powder’)

šulud ‘pass underneath ’

(√šul ‘be in, be under’)

təǰəd ‘roll ’

(√təč ‘roll off, tumble down’)

t’agʷtəd ‘put on top’

(√t’agʷt ‘be on top’)

t’uc’ud ‘shoot  (target)’

(√t’uc’ ‘be shot, fired on’)

x&alad ‘write ’

(√x&al ‘be written’)

xʷəxʷaʔxʷaʔəd ‘make  lighter’

(√xʷəxʷaʔxʷaʔ ‘be lightweight’)

Table 22: Internal causative stems formed from free radicals

In addition to transparent forms such as these, there are a number of more or less idiomatic

expressions that seem to follow the internal causative pattern such as dᶻək’ʷud ‘lead astray,

mislead’ (from √dᶻək’ʷ ‘travel, wander’), x&aʔx&aʔəd ‘forbid someone’(√x&aʔx&aʔ ‘be powerful, be

taboo’), dukʷud ‘change something; bewitch someone’ (√dukʷ ‘be a-normal (e.g.,

supernatural)’), ƛ’əld ‘keep something still’ (√ƛ’əl ‘be silent’), and q’pud ‘gather up’(√q’əp

‘form a lump; cramp up (muscle)’. There is also at least one internal causative form in the texts

which is only attested in the reflexive — dᶻəgʷaʔcut ‘smarten up, be clever’, based on √dᶻəgʷaʔ

‘be good at, have talent for’. This form also seems to have a rather idiomatic — though not

entirely opaque — meaning.

There is also a very large set of internal causative stems based on bound radicals, a number

of which are given in Table 23:

ʔabəd ‘extend , give ’

(º√ʔab ‘be extended’; cf. ʔabačiʔb ‘extend hand’)

ʔay’waʔsəd ‘exchange ’

(from º√ʔay’ ‘be traded’ + -waʔs ‘pair’)

bəq’əd ‘swallow ’

(º√bəq’ ‘have in mouth’; cf. bəq’dxʷ ‘taste ’))

biƛ’id ‘smash ’

(º√biƛ’ ‘be smashed, be crumbled’; cf. biƛ’il ‘crumble’)

c’əld ‘defeat ’

(º√c’əl ‘be defeated’; cf. c’əldxʷ ‘manage to defeat ’)

c’uqʷud ‘suck on ’

(º√c’uqʷ ‘be sucked on’; cf. c’uqʷačiʔb ‘suck on hand’)

č’aʔəd ‘dig for  (roots)’

(º√č’aʔ ‘be dug up’; cf. č’aʔəb ‘dig for  (roots)’)

č’ədᶻəd ‘sneak up on ’

(º√č’ədᶻ ‘be stalked’; cf. č’ədᶻalikʷ ‘stalk prey’)

digʷid ‘advise ’

(º√dikʷ ‘take advice’; cf. xʷdikʷ ‘advice’)

gʷəč’əd ‘look for ’

(º√gʷəč’ ‘be sought’; cf. gʷəč’əb ‘seek  for self’)

gʷəlald ‘punish , kill ’

(º√gʷəlal ‘be hurt’; cf. gʷəlalalikʷ ‘kill , slaughter ’)

kʷaʔd ‘let go of ’

(º√kʷaʔ ‘be released’; cf. kʷaʔdxʷ ‘manage to release ’)

kʷaxʷad ‘help ’

(º√kʷaxʷ ‘be helped’; cf. kʷaxʷdxʷ ‘manage to help ’)

k’ʷalč’əd ‘bend  backwards’

(º√k’ʷalč’ ‘be bent backwards’; cf. k’ʷalč’əb ‘bend self backwards’)

laʔəd ‘locate ’

(º√laʔ ‘be located’; cf. laʔyid ‘locate  for ’)

ɬuq’ʷud ‘peel ’

(º√ɬuq’ʷ ‘be peeled’; cf. ɬuq’ʷač ‘be bald (lit. ‘peeled head’)’)

ƛ’agʷəd ‘stitch  (mat)’

(º√ƛ’akʷ ‘be stitched’; cf. ƛ’agʷəb ‘make mats’)

ƛ’ip’id ‘squeeze ’

(º√ƛ’ip’ ‘be compressed’; cf. dxʷƛ’ip’alikʷ ‘player in hand game’)

p’ic’id ‘wring  out’

(º√p’ic’ ‘be wrung out’; cf. p’ic’albixʷ ‘milk (cow)’)

p’t’ad ‘store , tidy ’

(º√p’t’ ‘be stored’; cf. p’t’alikʷ ‘save ’)

q’p’ud ‘pay ’

(º√q’p’ ‘be compensated’; cf. q’p’alikʷtxʷ ‘pay for a crime’)

taʔəd ‘take  over there’

(º√taʔ ‘be in place’; cf. ʔəstaʔtxʷ ‘have  in place’)

tqad ‘close , block  off’

(º√tq ‘be closed’; cf. təqdxʷ ‘block ’s path’)

tx&ʷud ‘pull on ’

(º√tx&ʷ ‘be pulled’; cf. təx&ʷgʷiɬ ‘pull canoe’)

t’ugʷud ‘figure  out’

(º√t’ukʷ ‘be measured’; cf. t’ukʷtəd ‘a tape measure’)

xʷacad ‘carry ’

(º√xʷac ‘be hoisted’; cf. xʷaʔxʷcəb ‘be rock-lifting’)

xʷəbəd ‘toss ’

(º√xʷəb ‘be thrown’; cf. xʷəbagʷil ‘throw oneself’)

xʷit’ild ‘lower ’

(º√xʷt’ ‘be fallen, be descended’; cf. xʷt’agʷil ‘climb down’)

x&ədəd ‘push ’

(º√x&əd ‘be pressed’; cf. x&ədačiʔb ‘push hands away’)

x&əqəd ‘wrap ’

(º√x&q ‘be wrapped, be tied’; cf. x&qagʷil ‘tie oneself down’)

x&ibid ‘grab  by throat’

(º√x&ib ‘be grabbed, clawed’; cf. x&ibigʷs ‘have one’s things in hand’)

x&ʷadᶻad ‘punish , annihilate ’

(º√x&ʷadᶻ ‘be injured’; cf. x&ʷadᶻalikʷ ‘slaughter ’)

Table 23: Internal causative stems formed from bound radicals

Although not attested as independent, patient-oriented forms such as that shown in (39a), each of

these radicals is attested as part of other stems and interacts with other valency-increasing

morphemes in a manner consistent with a monovalent patient-oriented radical.

In addition to synchronically analyzable stems containing the internal causative, there are

also a large number of transitive verbs that appear to contain either -t or -š but which are not

based on clearly-attested radicals found in other verb forms. Several of these are given in Table

24:

bak’ʷad ‘move  quickly’

(*√bak’ʷ ‘move quickly’)

baɬad ‘cure  with shamanism’

(*√baɬ ‘be cured by shamanism’)

biq’id ‘press ’

(*√biq’ ‘be pressed down’)

bisəd ‘select ’

(*√bis ‘be selected’)

čədᶻq’ʷəd ‘rub  together’

(*√čədᶻq’ʷ ‘be rubbed together’)

čətx&ʷəd ‘gobble  up’

(*√čətx&ʷ ‘be gobbled up’)

dᶻilid ‘despise ’

(*√dᶻili ‘be despised’)

gəq’əd ‘open something’

(*√gəq’ ‘be opened’)

hilid ‘command ’

(*√hil ‘obey’)

ɬalš ‘remove  from fire’

(*√ɬalš ‘removed from fire’)

ƛ’aq’ʷad ‘lie in wait for ’

(*√ƛ’aq’ʷ ‘be ambushed’)

pačad ‘lay  out’

(*√pač ‘be laid out’)

paq’ad ‘distribute ’

(*√paq’ ‘be scattered, distributed’)

qʷaɬad ‘drive  (animal); drive  off’

(*√qʷaɬ ‘be driven off’)

q’x&ad ‘insult ’

(*√q’x& ‘be insulted’)

sax&ad ‘scrape ’

(*√sax& ‘be scraped’)

sət’əd ‘lift ’

(*√sət’ ‘be raised’)

təbaš ‘crave ’

(*√təb ‘have a craving’)

xʷalusəd ‘wave ’

(*√xʷalus ‘be waved’)

x&ʷuƛ’ud ‘chew  up’

(*√x&ʷuƛ’ ‘be chewed up’)

Table 24: Inherently transitive stems containing the internal causative

Aside from analogy with the forms in Table 22 and Table 23, the diachronic presence of the

internal causative can also be inferred from the allomorphy shown by the stem-final /d/:

(45) a. huy qʷaɬadəxʷ sixʷ tiʔəʔ wiw’su ɬulilcutəs

huy qʷaɬad=əxʷ sixʷ tiʔəʔ wiw’su ɬu=lil–t–sut=əs

SCONJ drive.off=now PTCL PROX chlidren IRR=far–ICS–REFL=3SBJ

‘so he drove the children off so they would get away’

(Hess 1998: 94, line 94)

b. gʷəl ʔuq%aɬatubuɬ

gʷəl ʔu–qʷaɬa–t–ubuɬ

SCONJ PFV–drive.off–ICS–1PL.OBJ

‘but he drove us off’

(Hess 1998: 93, line 77)

c. kʷaʔ x%ʷul’əxʷ əlgʷəʔ ləcubiq’id dxʷʔal tiʔəʔ hud

kʷaʔ x%ʷul’=əxʷ əlgʷəʔ ləcu–biq’i–d dxʷ–ʔal tiʔəʔ hud

PTCL only=now PL PROG.STAT–press–ICS CNTRPT–at PROX fire

‘it seems they just pressed [her] down into the fire’

[MS Basket Ogress, line 67]

d. gʷəl biq’itəbəxʷ ʔə tiʔəʔ wiw’su dxʷʔal sx%ʷul’səxʷ ʔuƛ’iq’ ti č’əƛ’č’ƛ’aʔ

gʷəl biq’i–t–əb=əxʷ ʔə tiʔəʔ wiw’su dxʷ–ʔal s=x%ʷul’=s=əxʷ

then press–ICS–PASS=now PR PROX children CNTRPT–at NM=only=3PO=now

ʔu–ƛ’iq’ ti č’əƛ’–č’ƛ’aʔ

PFV–emerging SPEC DSTR–stone

‘then [she] was pressed down by the children so that only the stones were sticking out’

[AJ Basket Ogress, line 109]

As shown by these examples, the final consonant in transitive stems like qʷaɬad ‘drive someone

off’ and biq’id ‘press down on something’ shows the same ultimate-final voicing alternation

pattern seen in internal causative forms as those in (43). Unlike bound radicals such as those in

Table 23, however, putative radicals such as *√qʷaɬ ‘be driven off’ and *√biq’ ‘be pressed

down’ are not found as part of other verb forms independent of the internal causative and so are

treated separately from these for the purposes of classification.

Although the majority of stems containing -t conform to the internal causative pattern

described above, there is a not-insignificant number of stems in which -t acts as some kind of

valency-increaser other than a causative. In the largest set of such forms, the effect of -t on the

radical is that of an applicative. The stems found to date that follow this pattern are given in

Table 25:

ʔilid ‘sing ’

(√ʔil ‘sing’)

ʔulud ‘sing to ’

(º√ʔul ‘sing’; cf. ʔuliʔɬ ‘sing lullaby’)

bəlx&ʷəd ‘pass ’

(√bəlx&ʷ ‘be beyond’)

cut ‘speak to ’

(√cut ‘speak’)

dᶻaqad ‘mourn ’

(º√dᶻaq ‘be in mourning’; cf. dᶻaqəbid ‘mourn for ’)

gʷiid ‘invite , call to ’

(√gʷi ‘make an invitation’)

gʷuhud ‘bark at ’

(º√gʷuh ‘bark (dog)’; cf. gʷuhəb ‘bark’)

k’ʷaɬad ‘examine ’

(º√k’ʷaɬ ‘look closely, peer’; cf. k’ʷək’ʷaɬəb ‘be near-sighted’)

k’ʷilid ‘peek at ’

(√k’ʷil ‘peek’)

luud ‘hear ’

(º√lu ‘hear’)

ɬild ‘give food to ’

(√ɬil ‘make a gift of food’)

q’əlsəd ‘steam ’

(√q’əls ‘cook with steam’)

šidᶻəd ‘attack  by stealth’

(√šidᶻ ‘launch sneak attack’)

wiliq’ʷid ‘ask of ’

(√wiliq’ʷ ‘make an enquiry’)

x&idid ‘growl at ’

(º√x&id ‘be growling’; cf. x&idib ‘growl’)

x&ʷaq’ʷad ‘be concerned about ’

(√x&ʷaq’ʷ ‘be worried, be preoccupied’)

Table 25: Applicative uses of -t

These verbs are based on monovalent intransitive radicals with agentive, rather than patientive,

subjects, and the affixation of -t adds a direct object rather than a subject:

(46) a. ʔiləxʷ tiʔəʔ qaw’qs

ʔil=əxʷ tiʔəʔ qaw’qs

sing=now PROX raven

‘now Raven sings’

(Hess 1998: 57, line 38)

b. ƛ’ubəxʷ ʔuʔilitəb ʔə tsiʔiɬ x%ənimulic’aʔ kʷi sqəlalituts

ƛ’ub=əxʷ ʔu–ʔili–t–əb ʔə tsiʔiɬ x%ənimulic’aʔ

okay=now PFV–sing–ICS–PASS PR DIST:FEM name.of.Crow

kʷi sqəlalitut–s

REM spirit.power–3PO

‘x&ənimulic’aʔ ought to sing to her spirit power’

(Hess 1998: 61, line 25)

The semantic role of the object varies according to the meaning of the base: verbs of speaking

(cut ‘speak to someone’, wiliq’ʷid ‘ask someone’) or speech-like actions (ʔulud ‘sing to

someone’, gʷuhud ‘bark at something’) add a HEARER, while other verbs add roles such as

MOTIVE (x&ʷaq’ʷad ‘be concerned about something’, dᶻaqad ‘mourn someone’), BENEFACTIVE/

MALEFACTIVE (ɬild ‘give food to someone’, šidᶻəd ‘attack someone by stealth’), or PERCEPT

(k’ʷilid ‘peer out at something from behind barrier’). In addition to the forms listed in Table 25,

there are a number of other verbs which appear to follow this pattern but which are based on

radicals that are not otherwise attested or easily analyzed in other forms. These include p’aʔəd

‘try something’ (from *√p’aʔ ‘be attempted’) and səgʷqəd ‘whisper to’ (*√səgʷq ‘whisper’).

There is also an attested radical, √taš ‘spin with leg spindle’, which has an idiomatic form, tašad

‘stroke something’, in which what appears to be the internal causative suffix follows the

applicative pattern.

The -t suffix also appears in a number of forms in which it acts simply as a transitivizer,

converting a bivalent intransitive base into a monotransitive verb. Verbs that follow this pattern

include ʔaladᶻiʔɬəd ‘babysit someone’ (from ʔaladᶻiʔɬ ‘babysit someone’[= º√ʔaladᶻ ‘care for

someone’ + -iʔɬ ‘child’]), čəbaʔəd ‘backpack something’ (from √čəbaʔ ‘be loaded down with

something’), ƛ’alš ‘wear something’ (√ƛ’al ‘put something on’), tabəd ‘do something’ (√tab

‘deal with’), ʔišɬš ‘paddle something [canoe]’ (√ʔišɬ ‘paddle canoe’), k’ʷič’id ‘butcher

something’ (√k’ʷič’ ‘butcher something’), and ʔuləx&əd ‘gather something’ (√ʔuləx& ‘forage for’).

This set of forms may also include x&aƛ’əd ‘favour someone’ (º√x&aƛ’ ‘be desirous of’) and təbaš

‘crave something’ (*√təb ‘have a craving’), although the bare radicals are not attested in the

present corpus, making it hard to ascertain their inherent valency. Another possible candidate for

a transitivizing use of the internal causative would be suxʷtəš ‘recognize someone’, though the

lack of tokens of this form makes it difficult to be sure what the underlying semantics of the

radical would be.

The internal causative suffix also appears to be part of certain more complex transitive stems

whose synchronic analysis is uncertain. These include c’əɬqiwsəd ‘cut something up’ and

xʷak’ʷabičəd ‘get someone dirty’. A few intransitive verbs appear to contain -t as well, based on

the shape of their apparent radicals or on voicing alternations in word-final position. These

include a small set of verbs for making noise (tukʷud ‘thump’, k’ʷx&ʷiqid ‘make noise’, and sx&ʷid

‘make swishing sound’), the verbs ƛ’čabəd ‘double self over’ and gʷəƛ’əlad ‘stop’, and the

bivalent intransitive verb ʔəɬəd ‘feed on’. Unlike the others in this group, this last form has an

attested bound radical, º√ʔəɬ ‘be eating’ found in other forms such as ʔəɬtxʷ ‘feed someone’ and

sʔəɬəd ‘food’; however, verb ʔəɬəd itself is intransitive, and so it is not possible to analyze this

form synchronically as containing the internal causative morpheme.

2.1.2.2 External causative -txʷ

The next most frequent valency-increasing affix in Lushootseed is -txʷ ‘external causative

[ECS]’. Like -t, this affix is a transitive suffix that is added (with a few exceptions) to monovalent

radicals to form a transitive stem expressing an event in which an AGENT causes a PATIENT or

THEME to come into the state expressed by the radical; however, the AGENT in -txʷ forms is

construed as being somehow less directly involved in or affected by the event than it is in stems

formed with the internal causative. As with any causative, the new argument is realized as the

syntactic subject and the erstwhile subject of the base is realized as a direct object:

(47) a. ʔuʔux%ʷ čəd

ʔu–ʔux%ʷ čəd

PFV–go 1SG.SUB

‘I went’

(Hess 1995: 6, ex. 1)

b. ʔuʔux%ʷtubš ti č’ač’as

ʔu–ʔux%ʷ–txʷ–bš ti č’ač’as

PFV–go–ECS–1SG.OBJ SPEC child

‘the boy took me’

(based on Hess 1995: 42)

c. ʔuʔux%ʷtub čəd ʔə ti č’ač’as

ʔu–ʔux%ʷ–txʷ–b čəd ʔə ti č’ač’as

PFV–go–ECS–PASS 1SG.SUB PR SPEC child

‘I was taken by the boy’

(Hess 1995: 33)

As shown in (47a) and (b), when -txʷ is added to a monovalent radical, it forms a transitive verb

in which the AGENT is the subject and the PATIENT or THEME is the direct object, much like the

stems formed with the internal causative, although the object of -txʷ forms is marked by the b-

According to Bates, Hess, and Hilbert (1994), the form gʷəƛ’əlad ‘stop’ is derived diachronically from the radical

√ƛ’əl ‘be silent’, and the verb ƛ’čabəd ‘double self over’ may be related to √čaba ‘add to’; however, the words

themselves are non-compositional enough to warrant treatment here as synchronically unanalyzable forms.

series, rather than the s-series, of object-markers (Section 8.1.2). The object of external causative

verbs is a true direct object and undergoes the usual direct-object centred syntactic processes

such as passivization (47c); note that in the passive, -txʷ becomes [-tu-], as it does in the

presence of the object-markers. The affix shows no other type of morphophonemic interactions

with its base or with other affixes. When the derived stem takes an overt NP argument, this

argument is interpreted as direct object:

(48) ʔuʔux%ʷtxʷ ti č’ač’as

ʔu–ʔux%ʷ–txʷ ti č’ač’as

PFV–poured–ECS SPEC water

‘s/he took the boy’

(Hess 1995: 22, ex. 3b)

This property of transitive verbs is discussed in more detail in Section 8.2.2 below. The

affixation of the external causative suffix does not trigger any alternations in its base, although it

does show some morphophonemic interaction with some following affixes, most notably the

passive suffix (Section 6.2) and the object- (8.1.2) and reflexive-markers (8.1.3), all of which

cause the final /xʷ/ of the external causative suffix to become /u/. When followed by the

reciprocal marker (8.1.4), the external causative is realized simply as /t/.

The forms in (47) are based on a radical expressing motion, ʔux&ʷ ‘go’. A wide range of such

radicals combine with -txʷ to form verbs of taking and bringing such as those given in Table 26:

ʔəƛ’txʷ ‘bring ’

(√ʔəƛ’ ‘come’)

ʔibəštxʷ ‘take  for a walk’

(√ʔibəš ‘travel, walk’)

ʔux&ʷtxʷ ‘take ’

(√ʔux&ʷ ‘go’)

čubətxʷ ‘take  ashore’

(√čubə ‘go inland’)

gʷahtxʷ ‘take  along’

(√gʷa ‘accompany, go along’)

gʷax&ʷtxʷ ‘take  for a walk’

(√gʷax&ʷ ‘take a stroll’)

kʷatačtxʷ ‘carry  up a hill’

(√kʷatač ‘climb’)

k’ʷit’txʷ ‘take  down to shore’

(√k’ʷit’ ‘go down to shore’)

ɬaʔtxʷ ‘bring  to a place’

(√ɬaʔ ‘arrive at a specific place’)

ɬaliltxʷ ‘bring  ashore’

(√ɬalil ‘go ashore’)

This form also means ‘make  travel’; with this reading, the stem belongs with the forms in Table 27 below.

This form is also attested in the speech of older speakers as čubəstxʷ, the -stxʷ version of the suffix being perhaps

an archaic form which is still attested in some other languages of the family.

This is the gloss of the word as it is used in Skagit. In Snohomish, this verb also applies to climbing trees and

ladders, while the Skagit use √ʔigʷəł for these latter two activities.

ɬčiltxʷ ‘arrive with ’

(√ɬčil ‘arrive’)

q’iltxʷ ‘take  by canoe’

(√q’il ‘be aboard’)

saq’ʷtxʷ ‘fly off with ; fly  (airplane)’

(√saq’ʷ ‘fly’)

saxʷəbtxʷ ‘run off with , kidnap ’

(√saxʷəb ‘jump, sprint’)

šədᶻaltxʷ ‘take  outside’

(√šədᶻal ‘go outside’)

təlawiltxʷ ‘run off with ’

(√təlawil ‘run’)

tuliltxʷ ‘take  across river’

(√tulil ‘cross river’)

t’uk’ʷtxʷ ‘take  home’

(√t’uk’ʷ ‘go home’)

Table 26: Verbs of taking and bringing formed with -txʷ

In such forms, the radical expresses the type of motion undergone by the THEME while the suffix

adds an AGENT responsible for causing that motion. Unlike the English translation equivalents of

many of these stems, there is no inherent telicity or notion of transfer expressed by these verbs

themselves, although these notions may be implied by context. As with all -txʷ forms, the stems

in Table 26 are transitive; if a recipient is involved, it may be expressed as an oblique object

introduced by the preposition dxʷʔal ‘towards’, as in (49a):

(49) a. ʔux%ʷtubəxʷ tiʔiɬ sʔuladxʷ dxʷʔal tiʔiɬ sčətxʷəd

ʔux%ʷ–txʷ–b=əxʷ tiʔiɬ sʔuladxʷ dxʷ–ʔal tiʔiɬ sč ətxʷəd

go–ECS–PASS=now DIST salmon CNTRPT–at DIST black.bear

‘the salmon was taken to Black Bear’

(Hess 1995: 154, line 67)

b. ləɬčil čəd, capaʔ, dxʷʔal dəgʷi

lə=ɬčil čəd capaʔ dxʷ–ʔal dəgʷi

PRG–arrive 1SG.SUB grandfather CNTRPT–at you

‘I am coming, Grandfather, to you’

(Hess 2006: 28, line 157)

As shown in (49b), the preposition is the same as that used to express the goal of motion in the

non-causativized forms.

In addition to verbs of taking and bringing, -txʷ is used to form a wide variety of transitive

verbs from intransitive stems:

ʔalalustxʷ ‘do to ’

(√ʔalalus ‘happen’)

ʔatxʷ ‘put there’

(√ʔa ‘be there’)

ʔəɬtxʷ ‘feed ’

(º√ʔəɬ ‘be eaten’; cf. ʔəɬəd ‘feed on ’)

ʔistaʔtxʷ ‘do the same to ’

(√ʔistaʔ ‘be the same’)

ʔup’txʷ ‘seat  on another’s lap; seat person on ’s lap’

(√ʔup’ ‘be seated on lap’

c&əbaʔtxʷ ‘pack  on one’s back’

(√čəbaʔ ‘be loaded down with ’)

dukʷtxʷ ‘get angry with ’

(√dukʷ ‘be anormal’)

gʷədiltxʷ ‘sit  down’

(from gʷədil ‘sit down’ based on √gʷəd ‘down’)

həliʔtxʷ ‘cure ’

(√həliʔ ‘be alive’)

hiwiltxʷ ‘go ahead with ’

(√hiwil ‘proceed’)

huygʷastxʷ ‘marry ’

(from √huy ‘be made’ + -gʷas ‘pair’)

kiistxʷ ‘stand  up’

(√kiis ‘stand up’)

lax&txʷ ‘remind ’

(√lax& ‘recall, remember’)

ɬidtxʷ ‘tie to ’

(√ɬid ‘be tied’)

ƛ’ax&ʷtxʷ ‘bring up , raise ’

(√ƛ’ax&ʷ ‘grow’)

ƛ’iq’ačiʔbtxʷ ‘make ’s hands sticky

(from √ƛ’iq’ ‘be sticky’ + -ačiʔ ‘hand’)

q’iltxʷ ‘put load into ’

(√q’il ‘be aboard’)’

saʔtxʷ ‘dislike , hate ’

(√saʔ ‘be bad’)

saq’ʷtxʷ ‘fly off with ’

(√saq’ʷ ‘fly’)

šəɬt’əbiɬədtxʷ ‘make rope of ’

(from √šəɬ ‘make ’ + √t’əbiɬəd ‘rope’)

šuɬtxʷ ‘show to ’

(√šuɬ ‘appear, be visible’)

tədᶻiltxʷ ‘put  to bed’

(√tədᶻil ‘go to bed, lie in bed’)

təɬtxʷ ‘make  true, speak truth’

(√təɬ ‘be true’)

t’ičibtxʷ ‘make  wade’

(√t’ičib ‘wade’)

t’uc’iltxʷ ‘fire ’

(t’uc’il ‘fire weapon’ from √t’uc’ ‘be shot’)

x&aʔx&aʔtxʷ ‘forbid ’

(√x&aʔx&aʔ ‘powerful, taboo’)

x&ilix&txʷ ‘make war on ’

(√x&ilix& ‘be at war’)

Table 27: Causative stems formed with -txʷ

As with verbs of taking and bringing, the stems here are formed on intransitive bases, and adding

-txʷ forms a transitive stem following the pattern illustrated in (47) above, with the new

argument taking the syntactic role of subject. The bulk of radicals in Table 27 fall roughly into

two groups — agent-oriented radicals with AGENT- or ACTOR-like subjects (corresponding to

what Gerdts 2006 classifies as “unergative” predicates in Halkomelem) which give rise to cross-

linguistically ordinary causative stems (√kiis ‘stand up’ > kiistxʷ ‘stand something up’, √t’ičib

‘wade’ > t’ičibtxʷ ‘make someone wade’), and radicals expressing subjective characterizations

or psychological states which give rise to causative stems with figurative rather than literal

meanings (√dukʷ ‘be a-normal’ > dukʷtxʷ ‘get angry with someone’ [lit. ‘make someone a-

normal (in one’s mind)’], √saʔ ‘be bad’ > saʔtxʷ ‘dislike something, hate something’ [lit. ‘make

something bad (in one’s mind)’], √x&aʔx&aʔ ‘powerful, taboo’ > x&aʔx&aʔtxʷ ‘forbid something’ [lit.

‘make something taboo’]).

There are also a few patient-oriented radicals in this set as well —

√həliʔ ‘be alive’ > həliʔtxʷ ‘cure something’, √ɬid ‘be tied’ > ɬidtxʷ ‘tie to something’ —

Also included in this group would be hikʷtxʷ ‘respect something’ [lit. ‘make someone big (in one’s mind)’,

derived from the adverb √hikʷ ‘big’.

although none of these is particularly high on the scale of semantic transitivity (Hopper &

Thompson 1980).

As noted by Hess & Bates (1998), there are a few stems where the direct object may express

some other semantic role than that expressed by the subject of the radical:

(50) a. q’iltxʷ ti adsƛ’əlayʔ

q’il–txʷ ti ad–sƛ’əlayʔ

be.aboard–ECS SPEC 2SG.PO–shovelnose.canoe

‘load your shovelnose canoe!’

(Hess & Bates 1998: 230, ex. 24)

b. ʔuʔup’txʷ tsi sɬadəyʔ

ʔu–ʔup’–txʷ tsi sɬadəyʔ

PFV–be.on.lap–ECS SPEC:FEM woman

‘s/he put [her] on the woman’s lap’

(Hess & Bates 1998: 231, ex. 25)

In (50a), the verb q’iltxʷ is built on the radical √q’il ‘be aboard conveyance’, whose subject is the

person or thing which is aboard the conveyance (most often a canoe); the direct object of the

transitive form q’iltxʷ, however, is the conveyance rather than the load or the rider. Similarly,

ʔup’txʷ, derived from √ʔup’ ‘be on a person’s lap’, can appear in clauses in which the person

whose lap is being sat upon is the direct object, as in (50b), although a more ordinary

government pattern is also found with this verb in which the direct object is the person being

made to sit, as show in the example in (58b) below.

In most of the examples in Table 27, the bases are monovalent radicals; however, in one or

two cases the causative is added to a bivalent intransitive base. The effect on the valency and

government pattern of the verb in these cases depends on the stem. For example, one of these

verbs, saq’ʷtxʷ ‘fly off with something; fly something [airplane]’, has two possible

interpretations, one following the causative pattern of other motion verbs shown in Table 26, the

other following a more general causative pattern. With other verbs such as čəbaʔtxʷ ‘pack

something on one’s back’ (from √čəbaʔ ‘be loaded down with something’) and t’uc’iltxʷ ‘fire

something’ (from t’uc’il ‘fire weapon’), the effect is to transitivize the verb:

(51) a. ləsčəbaʔ ʔə tə hud

ləs–čəbaʔ ʔə tə hud

CONT–be.packing PR NSPEC wood

‘she’s loaded down with wood’

(Bates, Hess & Hilbert 1994: 61)

b. gʷəl ʔabil’əxʷ ɬučəbaʔtxʷ əlgʷəʔ tiʔəʔ dsxʷiʔxʷiʔ

gʷəl ʔabil’=əxʷ ɬu–čəbaʔ–txʷ əlgʷəʔ tiʔəʔ d–sxʷiʔxʷiʔ

then perhaps=now IRR–be.packing–ECS PL PROX 1SG.PO–game

‘well then perhaps they can backpack my game’

[MW Star Child, line 76]

Here, rather than adding an argument, the external causative promotes an oblique object to direct

object. The semantic notion of causation is still inherent in the meaning of the -txʷ form itself

(čəbaʔtxʷ ‘backpack something’ ≈ ‘cause oneself to be loaded down with something’).

In a few other cases, -txʷ not only transitivizes a bivalent intransitive base but also “shuffles”

the basic diathesis of the verb:

(52) a. hay, qədbaxʷ ʔə tiʔəʔ sdukʷ sč’ətxʷ

hay qəd–b=axʷ ʔə tiʔəʔ sdukʷ sč’ətxʷ

SCONJ fornicate–MD=now PR PROX low.life kingfisher

‘so, she [Helldiver] has adulterous sex with that low-life Kingfisher’

(Hess 2006: 21, line 243)

b. gʷəl huy qədəbtxʷaxʷ tsiʔəʔ čəgʷas ʔə tiʔəʔ sbəq’ʷaʔ, tsiʔəʔ x%ʷuʔx%ʷəyʔ

gʷəl huy qəd–b–txʷ=axʷ tsiʔəʔ čəgʷas ʔə

then SCONJ fornicate–MD–ECS–now PROX:FEM wife PR

tiʔəʔ sbəq’ʷaʔ tsiʔəʔ x%ʷuʔx%ʷəyʔ

PROX heron PROX:FEM helldiver

‘and so then he [Kingfisher] seduces the wife of Heron, Helldiver’

(Hess 2006: 12, line 45)

(52a) shows the middle form qədəb ‘have illicit sex with someone, commit adultery with

someone’ which takes as its subject the expression of the adulterer (that is, the married person

who cheats on their spouse) and as an oblique object the expression of the person with whom

they cheat. In the causative form, qədəbtxʷ ‘seduce into adultery’, the adulterer’s sexual partner

— the seducer — is the subject and the adulterer is a direct object. Thus, the verb is transitivized

but the subject of the intransitive stem becomes the direct object of the transitive verb and the

oblique object of the intransitive stem becomes the subject. Rather than being a regular or

predictable syntactic operation, however, in this case the source of the change in government

pattern is the nature of the event: the proximate cause of adulterous behaviour is (attraction to)

the sexual partner, so it follows that in the causative form of this particular verb this event-

participant would be expressed as the causer/subject, resulting in an idiosyncratic, lexicalized

government pattern. Similar effects of the particular meanings of certain radicals are found with

a few other -txʷ forms.

In addition to verbs such as qədəbtxʷ ‘seduce into adultery’, whose semantic import seems

clearly to be causative but whose syntactic effects are somewhat idiosyncratic, there is another

group of stems that contain -txʷ but do not seem to be causative in either the semantic or

syntactic sense of the word. One small group of high-frequency verbs that fit this description are

verbs of speech in which -txʷ functions as an applicative rather than a causative (Hess & Bates

2004), adding a new argument realized as object rather than subject:

gʷaagatxʷ ‘speak to ’

(√gʷaagad ‘speak’)

qʷiʔadtxʷ ‘call  out’

(√qʷiʔad ‘holler, yell’)

tatabtxʷ ‘talk to ’

(º√tatab ‘speak’; cf. tatabəd ‘confer about ’)

t’ilibtxʷ ‘sing to ’

(√t’ilib ‘sing’)

x&ayəbtxʷ ‘smile at ’

(√x&ayəb ‘laugh’)

yəcəbtxʷ ‘tell  to ’

(yəcəb ‘report on ’ from º√yəc ‘report)

yəhubtxʷ ‘recite legend for ’

(º√yəhub ‘tell legend’; cf. syəhub ‘myth, legend’)

Table 28: Applicative uses of -txʷ

With most of these verbs, the effect of -txʷ seems to be to increase the valency of the stem by

adding a direct object with the semantic role of HEARER or PERCEIVER:

(53) a. huy, yəcəbaxʷ ʔə tiʔiɬ bibščəb ʔi tiʔiɬ suʔsuq’ʷaʔs, tətyika

huy yəcəb=axʷ ʔə tiʔiɬ bi–bščəb ʔi tiʔiɬ suʔsuq’ʷaʔ–s

SCONJ report=now PR DIST ATTN–mink and DIST younger.cousin–3PO

tətyika

Tutyika

‘he told of Little Mink and his younger cousin, Tutyika’

(Hess 1995: 141, line 42)

b. yəcəbtxʷaxʷ tiʔəʔ stubš ʔə ti sdaʔ ʔə tsiʔəʔ

yəcəb–txʷ=axʷ tiʔəʔ stubš ʔə ti sdaʔ ʔə tsiʔəʔ

report–ECS=now PROX man PR SPEC name PR PROX:FEM

‘he told the man her name’

[HM Star Child, line 116]

As shown in (53a), the verb yəcəb ‘report on something’ is a bivalent intransitive stem which

takes as its subject the speaker and expresses the topic of speech as an oblique object. As shown

by (53b), the -txʷ form of the verb is transitive and continues to express the topic of speech as an

oblique object.

In addition to being affixed to a wide range of verbal radicals, the external causative is also

found attached to a variety of other types of bases such as personal pronouns, the negative

adverb xʷiʔ, and interrogative words. In the case of personal pronouns, the effect of adding -txʷ

to a pronoun meaning ‘X’ is to create a verb meaning ‘make/let it be X’, as in (54):

It should be noted that for at least one of these stems, t’ilibtxʷ, there are two possible readings — ‘make someone

sing; turn something (radio) on’ and ‘sing to someone’:

(i) a. ʔut’ilibtxʷ čəd tiʔiɬ tidtid

ʔu–t’ilib–txʷ čəd tiʔiɬ tidtid

PFV–sing–ECS 1SG.SUB DIST radio

‘I turned on the radio’

(Hess & Bates 1998: 226, ex. 10)

b. tuləcut’ilibtubš

tu=ləcu–t’ilib–txʷ–bš

PAST=CONT–sing–ECS–1SG.OBJ

‘s/he was singing to me’

(Hess & Bates 1998: 232, ex. 30)

Similar ambiguities have not been reported for the other forms in Table 28.

(54) a. ʔəca kʷi ɬuʔabyid

ʔəca kʷi ɬu=ʔab–yi–d

I REM IRR=extend–DAT–ICS

‘the one who will give it to him [is] me’

b. ʔəcatxʷ kʷi ʔuʔabyid

ʔəca–txʷ kʷi ʔu–ʔab–yi–d

I–ECS REM PFV–extend–DAT–ICS

‘let the one who gives it to him be me!’

(based on Hess 1967a: 15)

Similarly, adding the external causative to the negative adverb creates a form with the literal

meaning ‘cause something not to be/not to happen’:

(55) xʷiʔtxʷ ləsaxʷəbtub

xʷiʔ–txʷ lə=saxʷəb–txʷ–b

NEG–ECS NEGP=run–ECS–PASS

‘do not let anyone kidnap him!’

(Hess 1967a: 15)

Note that syntax of the derived verb is the same as the syntax of the negative adverb, discussed

in Section 8.5 below.

The external causative is also associated with the interrogative word ʔəx&id ‘what happened?’,

forming another question word ʔəx&idtxʷ ‘do what to?’:

(56) a. ɬuʔəx%idəxʷ kʷi huy

ɬu=ʔəx%id=əxʷ kʷi huy

IRR=what.happened=now REM be.done

‘what can be done?’

(Hess 1998: 100, line 255)

b. ʔəsʔəx%idtxʷ čəxʷ sixʷ tsi adbədaʔ

ʔəs–ʔəx%id–txʷ čəxʷ sixʷ tsi ad–bədaʔ

STAT–what.happened–ECS 2SG.SUB PTCL SPEC:FEM 2SG.PO–offspring

‘what have you done to your daughter?’

(Hess 1998: 99, line 230)

The question word ʔəx&id is used, as in (56a), to ask about events and actions, formulating

questions glossed variously as ‘what happened?’ or ‘what was done?’. Its causative form,

ʔəx&idtxʷ, retains its interrogative sense but is syntactically a transitive verb, whose gloss might

be paraphrased as ‘cause what to happen to?’. Both in semantic and syntactic terms, this use of

-txʷ is perfectly regular and, given that interrogative words in Lushootseed are verbs, its

combination with ʔəx&id is unsurprising.

Another specialized use of -txʷ is its combination with the bound radical º√x&aƛ’ ‘be desirous

of’ to form complex desiderative expressions, as in (57):

(57) x%aƛ’tub ʔə tiʔəʔ sqʷəbqʷəbayʔ kʷi gʷəsx%ʷuƛ’utəbs

x%aƛ’–txʷ–b ʔə tiʔəʔ sqʷəb–qʷəbayʔ kʷi gʷə=s=x%ʷuƛ’u–t–b=s

desire–ECS–PASS PR PROX DSTR–dog REM SBJ=NM=chew.up–ICS–PASS=3PO

‘his being chewed up was desired by the dogs’

(Hess 1998: 46, ex. 1)

In such expressions, the form x&aƛ’txʷ ‘want that’ is used as a transitive verb whose subject is the

“wanter” and the direct object is the object of desire. In (57) the verb has been passivized and the

subject is a sentential nominal (Section 7.4.2.1) expressing a desired event. Further discussion of

desiderative expressions can be found in Sections 0 (on morphological desideratives) and 8.5.

The final aspect of the external causative to be discussed here concerns its overlap with the

internal causative, -t. As noted earlier, the basic semantic distinction between these two

causatives is the relative involvement of the AGENT in the event, state, or process expressed by

the radical (Beck 1996). With the internal causative, the AGENT is considered to be a more

integrated participant in the event either by dint of direct physical contact with the PATIENT or

greater affectedness of the AGENT (or, occasionally, the PATIENT) by the event itself. Many

external causative forms, like causatives in many other languages, imply that the change-of-state

or process undergone by the PATIENT was caused by a separate (often unspecified) action or

event initiated by the AGENT rather than being the result of direct action of the AGENT on the

PATIENT itself. As a result, the bulk of radicals select either the internal or the external causative,

depending on the nature of the state or process they express;

however, there are a number of

Note that, because transitive stems are derived from intransitive radicals, there is no potential for the formation of

causatives of transitive stems. Except for secondary suffixes (Section 2.1.3), Lushootseed disallows combinations of

valency-increasing suffixes, thereby excluding the possibility of adding a causative suffix to a transitive stem (which

would itself have to have been derived using a causative or applicative suffix).

radicals that have both an internal and an external causative form. Some of these are given in

Table 29:

ʔatxʷ ‘cause  to be somewhere’

ʔaʔəd ‘put ’

ʔup’txʷ ‘seat  on another’s lap’

ʔup’ud ‘seat  on one’s lap’

čaʔkʷtxʷ

‘take  out to sea’

čagʷəd ‘take  out to sea’

čaltxʷ

‘catch ’

čalad

‘chase ’

čəbaʔtxʷ ‘backpack ’

čəbaʔəd ‘backpack ’

daʔtxʷ ‘name  (spirit power)’

daʔad ‘name ’

dukʷtxʷ ‘make  angry/disgusted’

dukʷud ‘change , transform ’

k’ʷiltxʷ ‘cause  to peer out’

k’ʷilid ‘peek at ’

hədiw’txʷ ‘bring  inside’

hədiw’d ‘put  inside’

ɬidtxʷ ‘tie to ’

ɬidid ‘tie  up’

ƛ’iq’ačiʔbtxʷ ‘make ’s hands sticky’

ƛ’iq’id ‘stick  on’

ƛ’ubtxʷ ‘get  fixed’

ƛ’ubəd ‘agree to ’

qiq’txʷ ‘confine ’

qiq’əd ‘confine ’

q’iltxʷ ‘take  by canoe’

q’ilid ‘load  aboard’

wiliq’ʷtxʷ ‘ask  on another’s behalf’

wiliq’ʷid ‘ask ’

sulatxʷ ‘bring  to centre of room’

sulad ‘put  in centre of room’

x&aʔx&aʔtxʷ ‘forbid  (act)’

x&aʔx&aʔəd ‘deny permission ’

Table 29: Radicals with both -txʷ and -t forms

In some cases, the glosses of the verb stems make the difference in meaning between the two

types of causative clear, as in the case of ʔup’ud ‘seat someone on one’s own lap’ vs. ʔup’txʷ

‘seat someone on another’s lap’, given in context in (58):

(58) a. ʔup’ud tsi dsuq’ʷaʔ

ʔup’u–d tsi d–suq’ʷaʔ

seated.on.lap–ICS SPEC:FEM 1SG.PO–younger.cousin

‘he seats my younger cousin on his lap’

b. ʔup’txʷ tsi dsuq’ʷaʔ

ʔup’–txʷ tsi d–suq’ʷaʔ

seated.on.lap–ECS SPEC:FEM 1SG.PO–younger.cousin

‘he seats my younger cousin on another person’s lap’

(Hess 1993: 120n)

Other cases, like hədiw’d ‘put inside’ vs. hədiw’txʷ ‘bring inside’ or sulad ‘put in centre of

room’ vs. sulatxʷ ‘bring to centre of room’ are less clear and, judging by their glosses, the forms

seem to be nearly synonymous — or at least to overlap greatly in their potential to be used to

describe particular events. There are also a few forms (e.g., daʔad ‘name’ vs. daʔtxʷ ‘name spirit

power’, čalad ‘chase’ vs. čaltxʷ ‘catch someone’) where the difference seems to be lexicalized,

at least to the extent that the distinctions seen in the attested uses of these words do not obviously

conform to the more regular semantic nuances expressed by other contrasting uses of these two

suffixes.

Another set of radicals also combines with both causatives, but the resulting stems show

restrictions on potential aspectual inflections. With these radicals the external causative forms

are largely restricted to the stative aspect and are ungrammatical with the perfective, while the

internal causative forms are ungrammatical in the stative aspect. Both the external and the

internal causative forms of these radicals are given in Table 30:

ʔəsbəčtxʷ ‘have  laid down’

bəčad ‘set  down’

ʔəscaq’txʷ ‘have  impaled’

caq’ad ‘spear ’

ʔəsciltxʷ ‘have  dished up’

cilid ‘support , dish  up’

ʔəscqʷuɬtxʷ ‘have  hung on post’

—; (√cqʷuɬ ‘post’)

ʔəsdᶻaq’txʷ ‘have sex with ’

dᶻaq’ad ‘fell ’

ʔəshudtxʷ ‘keep  (fire) burning’

hudud ‘burn ’

ʔəshuytxʷ ‘have  prepared’

huyud ‘make , complete ’

ʔəsǰiq’txʷ ‘have  immersed’

ǰiq’id ‘immerse ’

ʔəskəkiʔtxʷ ‘have  in cradleboard’

—; (√kəkiʔ ‘cradleboard’)

ʔəsɬič’txʷ ‘have  cut into pieces’

ɬič’id ‘cut  with knife’

ʔəsƛ’altxʷ ‘be wearing ’

ƛ’alš ‘put  on’

ʔəsqʷatxʷ ‘have  laid out’

qʷatad ‘lay  down’

ʔəsqʷibtxʷ ‘have  ready’

qʷibid ‘prepare ’

ʔəstaʔtxʷ ‘have  in place’

taʔəd ‘put  in position’

ʔəst’agʷtxʷ ‘rest on top of ’

t’agʷtəd ‘put  on top’

ʔəsxʷəctxʷ ‘have  taken off/away’

xʷəcəd ‘take  (clothing) off’

ʔəsx&ək’ʷtxʷ ‘have  overturned’

x&k’ʷəd ‘turn  over’

ʔəsx&əqyalustxʷ ‘have ends of  wrapped’

x&əqəd ‘wind string or cloth around ’

ʔəsx&qaličtxʷ ‘have  packed up’

x&əqəd ‘wind string or cloth around ’

Table 30: External causative forms requiring the stative aspect

In most cases, the basic meaning of the verb stems in the two columns in Table 30 are the same,

the exceptions being the idiomatic expression ʔəsdᶻaq’txʷ ‘have sex with’ (from º√dᶻaq’

‘toppled, fallen’) and two forms based on nouns (ʔəskəkiʔtxʷ ‘have in cradleboard’ from √kəkiʔ

‘cradleboard’ and ʔəscqʷuɬtxʷ ‘have hung on post’ from √cqʷuɬ ‘post’). In the remainder of

cases the semantic distinction between the forms is purely aspectual, the combination of the

stative aspect and the external causative giving the reading of ‘have X in the state of’ as opposed

to the internal causative which has the usual meaning ‘cause X to be in the state of’. Of the two

forms, the internal causative form seems to be the more widely distributed and, aside from the

constraint against appearing in the perfective aspect, seems to be amenable to most other

aspectual inflections, including the imperfective (59a) and the progressive (59b):

(59) a. bəčatəbaxʷ tiʔiɬ k’ʷat’aq dəxʷʔibəš ʔə tiʔiɬ bibščəb ʔi tiʔiɬ suʔsuq’ʷas

Ø–bəča–t–əb–axʷ tiʔiɬ k’ʷat’aq dəxʷ=ʔibəš ʔə tiʔiɬ bi–bščəb

IMPF–be.lying–ICS–PASS–now DIST mat ADNM=travel PR DIST ATTN–mink

ʔi tiʔiɬ suʔ–suq’ʷa–s

and DIST ATTN–younger.cousin–3PO

‘a cat-tail mat is laid down for Little Mink and his younger cousin to walk on’

(Hess 1995: 142, line 44)

b. gʷəl ləbəčatəb ʔal tiʔiɬ qʷuʔ

gʷəl lə=bəča–t–əb ʔal tiʔiɬ qʷuʔ

then PROG=be.lying–ICS–PASS at DIST water

‘then it was being set down in the water’

(Hess 2006: 50, line 204)

It should be noted, however, that although all the forms in the left column of Table 30 are given

in the Lushootseed Dictionary in their citation form with the stative prefix, and the majority of

them are not attested in the present corpus in other aspects, two of them do appear in texts in

aspects other than the stative. These are huytxʷ ‘have something prepared’, which appears in the

imperfective aspect (60a), and x&qaličtxʷ ‘have something packed up’, which is used in the

continuous aspect (60b):

(60) a. bəhuytxʷ

bə=Ø–huy–txʷ

ADD=IMPF–be.done–ECS

‘he prepared it some more’

(Hess 2006: 47, line 137)

b. ɬuhuyud čəɬ čɬa ləsx%qaličtxʷ čxʷa ɬut’uk’ʷtxʷ

ɬu=huyu–d čəɬ čɬa ləs–x%q–alič–txʷ čxʷa

IRR=be.done–ICS 1PL.SUB 1PL.COORD CONT–wrapped–bundle–ECS 2SG.COORD

ɬu=t’uk’ʷ–txʷ

IRR=go.home–ECS

‘we will prepare it, we are bundling it up and you will take it home’

(Hess 1998: 80, line 68)

Thus, it may be that these and other -txʷ forms that seem to be restricted to the stative aspect

may in fact potentially appear in other (non-perfective) aspects as well, and their overwhelming

preference for the stative aspect may simply reflect pragmatic factors governing their usage

rather than formal aspects of their semantics.

2.1.2.3 Diminished control causative -dxʷ

Another frequent valency-increasing suffix is -dxʷ ‘diminished control [DC]’. This causative

transitive affix is added to intransitive stems and, like -t and -txʷ, adds an AGENT/subject to the

expression. However, in -dxʷ forms the AGENT is in less than complete control of the situation.

Consider the forms in (61), based on the radical √bəč ‘be lying down, be fallen from standing’:

(61) a. gʷəl ʔəsbəč, ʔəsx%ax%aqšəd ʔal kʷi ʔudəgʷiɬ ʔə tiʔiɬ qəlalgʷiɬ

gʷəl ʔəs–bəč ʔəs–x%a–x%aq•šəd ʔal kʷi ʔu–dəgʷ•gʷiɬ

SCONJ STAT–lie STAT–DSTR–wrapped•leg at REM PFV–middle•canoe

ʔə tiʔiɬ qəl•al•gʷiɬ

PR DIST bad•CNN•canoe

‘then he lay there with his feet wrapped in the middle of that funeral canoe’

[ML Mink and Tutyika I, line 80]

b. ʔubəčdubš ti sqʷəbayʔ

ʔu–bəč–dxʷ–bš ti sqʷəbayʔ

PFV–lying–DC–1SG.OBJ SPEC dog

‘the dog accidentally knocked me over’

(Hess 1995: 41, ex. 4b)

c. ʔubəčdub čəd ʔə ti sqʷəbayʔ

ʔu–bəč–dxʷ–b čəd ʔə ti sqʷəbayʔ

PFV–lying–DC–PASS 1SG.SUB PR SPEC dog

‘I was accidentally knocked over by the dog’

(Hess 1995: 41, ex. 4a)

In (61a), the bare radical is shown with its basic meaning, ‘be lying down’. The addition of the

diminished control suffix creates a verb meaning ‘knock something over’ — that is, ‘accidentally

cause something to be lying down’. Thus, the diminished control suffix adds an AGENT to the

stems with ICS and DC: 16/22; with ECS and DC: 4/22; with both: 2/22. Show the agent in ICS and DC more

alike

expression. This AGENT is expressed as a syntactic subject and the PATIENT (the THEME of the

radical) is expressed as a direct object. When the PATIENT is first- or second-person, the verb

takes b-form object markers (Section 8.1.2), as shown in (61b). The new object is treated

syntactically like any direct object, and is subject to object-centred syntactic processes such as

passivization (61c). In the passive and in the presence of the object-markers, -dxʷ becomes [-du-

]. When the derived stem takes an overt NP argument, this argument is interpreted as direct

object:

(62) ʔuk’ʷəɬdxʷ ti qʷuʔ

ʔu–k’ʷəɬ–dxʷ ti qʷuʔ

PFV–poured–DC SPEC water

‘s/he spilled the water’

(Hess 1995: 18, ex. 1b)

This property of transitive verbs is discussed in more detail in Section 8.2.2 below.

Like the external causative -txʷ, -dxʷ shows some morphophonemic interaction with certain

affixes that follow it, most notable the passive suffix (Section 6.2) and the object- (8.1.2) and

reciprocal-markers (8.1.3), all of which cause the final /xʷ/ of the diminished control suffix to

become /u/. When followed by the reciprocal marker (8.1.4), the diminished control suffix is

realized simply as /d/. Unlike -txʷ, however, -dxʷ triggers schwa epenthesis when it follows a

voiceless consonant and is word-final, as in the examples in (63):

(63) lax$dxʷ ‘remember’ < º√lax ‘remember’

t’uc’$dxʷ ‘manage to shoot’ < √t’uc’ ‘be shot’

(Hess 1967a: 12)

This schwa, however, is not recognized in Lushootseed orthography and is not included in any of

the published texts, nor will it be included in the examples presented in the remainder of this

grammar.

Table 31 gives a number of examples of stems formed with the diminished control

morpheme, along with the gloss provided for them in the source:

ʔadᶻqdxʷ ‘happen to meet up with ’

(º√ʔadᶻq ‘be met’; cf. ʔadᶻqbid ‘meet ’)

ʔaʔildxʷ ‘manage to put  there’

(ʔaʔil ‘get there’ from √ʔa ‘be there’)

ʔuq’ʷdxʷ ‘be left open to , be vulnerable to ’

(º√ʔuq’ʷ ‘be unplugged’; cf. ʔuq’ʷud ‘unplug ’)

bək’ʷdxʷ ‘manage to get all of ’

(√bək’ʷ ‘all, completely’)

caq’šadxʷ ‘lead ’

(caq’šad ‘step’ from √caq’ ‘be speared’ + -šad ‘leg’)

c’əldxʷ ‘manage to defeat ’

(º√c’əl ‘be defeated’; cf. c’əld ‘defeat ’)

čaldxʷ ‘catch up to ’

(√čal ‘be overtaken’

č’axʷdxʷ ‘manage to hit with a stick’

(√č’axʷ ‘be hit with a stick’)

dikʷdxʷ ‘instruct ’

(º√dikʷ ‘be advised’; cf. dxʷdigʷid ‘advise ’)

dukʷildxʷ ‘be dissatisfied with ’

(dukʷil ‘become strange’ from √dukʷ ‘be anormal’)

həliʔdxʷ ‘save the life of ’

(√həliʔ ‘be alive’)

huydxʷ ‘manage to do ’

(√huy ‘be done, be made, be finished’)

k’awdxʷ ‘collide with ’

(º√k’aw ‘be bumped’; cf. k’awqid ‘bump head’)

kʷaʔdxʷ ‘manage to let go of ’

(º√kʷaʔ ‘be released’; kʷaʔd ‘release ’)

kʷaxʷdxʷ ‘manage to help ’

(º√kʷaxʷ ‘be helped’; cf. kʷaxʷad ‘help ’)

kʷədxʷ ‘manage to take ’

(√kʷəd ‘be held, be taken’)

k’ʷəɬdxʷ ‘spill ’

(√k’ʷəɬ ‘pour out, spill out’)

labdxʷ ‘see ’

(√lab ‘appear’)

lax&dxʷ ‘remember ’

(√lax& ‘recall, remember’)

lək’ʷdxʷ ‘manage to eat ’

(√lək’ʷ ‘be eaten’)

ləxʷdxʷ ‘manage to stab ’

(º√ləxʷ ‘be stabbed, be cut’; cf. ləxʷud ‘stab ’

lildxʷ ‘draw away from ’

(√lil ‘far’)

ƛ’ubildxʷ ‘manage to improve ’

(√ƛ’ubil ‘improve’ from √ƛ’ub ‘good, well’)

pusdxʷ ‘manage to hit  with missile’

(√pus ‘be hit by  (missile)’)

p’alildxʷ ‘revive ’

(p’alil ‘regain consciousness’)

qəɬdxʷ ‘accidentally awaken ’

(√qəɬ ‘be awake’)

šudxʷ ‘catch sight of ’

(√šuɬ ‘appear, be visible’)

təqdxʷ ‘block ’s path’

(º√tq ‘closed’; cf. tqad ‘close , block  off’)

x&əɬdxʷ ‘injure ’

(√x&əɬ ‘be sick’)

x&ʷal’dxʷ ‘get the better of ’

(√x&ʷal’ ‘be unable, fail, lose’)

Table 31: Stems formed with -dxʷ

As can be seen in the glosses of many of these examples, the diminished control expressed by

-dxʷ generally has one of two sources — either the action is performed accidentally (qəɬdxʷ

‘accidentally awaken’, k’ʷəɬdxʷ ‘spill’), or the action is performed with some difficulty (c’əldxʷ

‘manage to defeat’, č’axʷdxʷ ‘manage to club, manage to get a lick in’). Which of the two types

of reading a verb will have depends loosely on the basic meaning of the radical. Thus, radicals

expressing non-desirable states or conditions unlikely to be desired by an actor (º√k’aw ‘be

bumped’, √k’ʷəɬ ‘pour out, spill out’) tend to have accidental readings, while radicals that

express more desirable states or the endpoints of willful action on the part of an AGENT (º√kʷaxʷ

See fn. 12 above.

This form is a bit odd as its meaning seems more closely related to the sense of √ dukʷ ‘be worthless’; however,

dukʷil is only attested with the meaning ‘become strange, become supernatural’ and not with the meaning ‘become

worthless’.

‘be helped’, ƛ’ubildxʷ ‘manage to improve’) tend to have the achieved-with-difficulty reading.

Similarly, radicals expressing undesirable states that might be resisted by a potential undergoer

(√c’əl ‘be defeated’, º√ləxʷ ‘be stabbed, be cut’) generally take -dxʷ with a reading of difficulty

in achievement. Some verbs of perception (labdxʷ ‘see’) and mental states (lax&dxʷ ‘remember’,

p’alildxʷ ‘bring around’) also take (or are only used with) -dxʷ, reflecting the lack of direct

conscious control we have over perceptual stimuli and mental processes.

However, with most

stems the source of the diminished control is ultimately context-dependent. Compare, for

example, the glosses of the sentences in (64):

(64) a. ʔuč’axʷdxʷ

ʔu–č’axʷ–dxʷ

PFV–clubbed–DC

‘he finally got a “lick” in [with his switch]’

b. ʔuč’axʷdubuɬ

ʔu–č’axʷ–dxʷ–buɬ

PFV–clubbed–DC–1PL.OBJ

‘he accidentally hit us with a stick’

(Bates, Hess & Hilbert 1994: 69)

Although the verb stems in the two sentences are the same, the glosses (based on the context of

utterance) are entirely different with respect to the locus of the diminished control. In the first

case, the AGENT is not in control due to the resistance of the PATIENT not wanting to be switched,

in the second case the diminished control comes from the inadvertent nature of the act. This type

of context-dependent localization of diminished control is extended even further in the sentence

in (65):

(65) gʷəhaw’əʔ x%ʷul’ ʔuɬəgʷəldxʷ tsiʔəʔ čəgʷas

gʷə=haw’əʔ x%ʷul’ ʔu–ɬəgʷɬ–dxʷ tsiʔəʔ čəgʷas–s

SBJ=PTCL only PFV–leave.behind–DC DIST:FEM wife–3PO

‘it would seem he only just left his wife behind’

(Hess 2006: 12, line 42)

Cf. ludxʷ ‘happen to hear , hear about ’ (from º√lu ‘hear’). In this form the diminished control suffix

functions as an applicative.

This sentence comes from a story in which Heron leaves his wife behind in their house (with no

great difficulty) to go fishing (deliberately) for a particular food that his wife, Little Diver, has

requested. The diminished control in this instance arises from the fact the Heron has no choice

but to leave his wife (who is feigning illness) behind — and that, when he does so, his wife’s

lover comes to visit her, making Heron’s diminished control of the situation the central point of

this section of the narrative. Thus, -dxʷ seems to be singularly unselective about the locus of

diminished control, requiring only that the AGENT not be in full control of some salient aspect of

the event under consideration.

There us a small number of radicals that form reflexive stems with -dxʷ:

hədʔiw’dubut

‘manage to get inside’

(√hədʔiw’ ‘be inside a house’)

ɬalildubut ‘manage to get ashore’

(√ɬalil ‘go ashore’)

šədᶻaldubut ‘manage to get outside’

(√šədᶻal ‘go outside’)

šulagʷildubut

‘manage to get in small space’

(šulagʷil ‘go in/under’ from √šul ‘be in, be under’)

təɬəɬdubut

‘manage to arrive safely’

(√təɬəɬ

‘arrive safely’)

Table 32: Reflexive stems formed with -dxʷ

For one of these stems, ɬalildxʷ ‘beach something’, the non-reflexive transitive diminished

control form is the only form attested; the only transitive forms attested for the other radicals in

this set are those formed with -txʷ ‘external causative’ (e.g., ɬaliltxʷ ‘bring ashore’). this may

stem from the fact that most of the AGENTs associated with -dxʷ forms are more akin to the

AGENTs associated with the internal causative, -t, and are treated as internal to rather than

external to the event being expressed by the stem, allowing for the use of the reflexive suffix,

which is otherwise only found with the internal causative form of verbs.

In addition to stems formed on identifiable roots, the diminished control suffix also appears

to be a part of a number of synchronically unanalyzable stems or stems based on putative

radicals which are not clearly attested as part of other forms. These include two high-frequency

verbs, haydxʷ ‘know’ and ʔəy’dxʷ ‘find’, as well as a few others such as qʷacdxʷ ‘doubt’. In

David Beck 10-2-7 2:19 PM

Comment: NOTE!!!!

these cases, the “presence” of the diminished control morpheme is seen in the transitivity of the

verb stem and by the fact that these verbs all take b-series objects:

(66) a. ʔuʔəy’dubuɬ čəxʷ ʔal tiʔəʔ ʔəƛ’ax%ad

ʔu–ʔəy’dxʷ–buɬ čəxʷ ʔal tiʔəʔ ʔəƛ’•ax%ad

PFV–find–1PL.OBJ 2SG.SUB at PROX come•side

‘you found us on this side’

(Hess 2006: 72, line 709)

b. ʔuhaydubš ti stubš

ʔu–haydxʷ–bš ti stubš

PFV–know–1SG.OBJ SPEC man

‘the man found out who I was’

(Bates, Hess & Hilbert 1994: 107)

This is fairly good evidence that diachronically these stems are formed from an intransitive

radical combined with the diminished control suffix; if the stem itself were transitive, the s-series

would be expected.

Finally, there are a few stems formed with -dxʷ from bivalent bases. When the base is

bivalent and intransitive, as in ƛ’aldxʷ ‘manage to getsomethingon’ (from √ƛ’al ‘put something

on’), the diminished control suffix acts as an applicative, promoting the oblique object of the

base to direct object, as well as adding the notion of diminished control. In the case of the

transitive radical √ɬəgʷɬ ‘leave something’, the base for ɬəgʷəldxʷ ‘manage to leave something

behind’, -dxʷ has no effect on the valency of the base, merely serving to mark diminished

control. Hess (1990) also notes two more stems formed with -dxʷ where this suffix does not

function as a causative. The first of these is pusil ‘throw something’, in which the diminished

control suffix functions merely as a syntactic transitivizer, promoting an oblique object of a

bivalent intransitive stem to direct object without affecting the semantic valency of the stem, as

shown in (67):

David Beck 10-2-7 2:19 PM

Comment: we may need to remove haydx

from this list, there is hayed and haytx in the

BO stories and hey’d in the reflexive and

reciprocal

Comment: actually, pus is bivalent

intransitive. you should take it out of the table

above and include it here as a radical

(67) a. ʔupusil ʔə tiʔəʔ č’ƛ’aʔ

ʔu–pus–il ʔə tiʔəʔ č’ƛ’aʔ

PFV–thrown–INCH PR PROX rock

‘s/he threw the rock’

(Hess 1990: 174, ex. 5)

b. ʔupusildxʷ tiʔəʔ č’ƛ’aʔ

ʔu–pus–il–dxʷ ʔə tiʔəʔ č’ƛ’aʔ

PFV–thrown–INCH–DC PR PROX rock

‘s/he threw the rock’

(Hess 1990: 174, ex. 6)

Here, the stem without -dxʷ — based on the radical √pus ‘be hit by something (missile)’ plus an

idiosyncratic use of the inchoative -il (Section 2.1.1.2) — is bivalent, taking the expression of

the missile as an oblique object (67a). When -dxʷ is added, the valency of the stem is unaltered

but the derived verb is transitive, expressing the missile as a direct object (67b). A similar pattern

in seen with the radical √q’il ‘be aboard’, as shown in (68):

(68) a. ʔuq’il dxʷʔal tiʔəʔ sdəxʷiɬ

ʔu–q’il dxʷ–ʔal tiʔəʔ sdəxʷiɬ

PFV–be.aboard CNTRPT–at PROX canoe

‘it is aboard the hunting canoe’

(Hess 1990: 174, ex. 7)

b. ʔuq’ildxʷ tiʔəʔ sdəxʷiɬ

ʔu–q’il–dxʷ tiʔəʔ sdəxʷiɬ

PFV–be.aboard–DC PROX canoe

‘s/he loaded the hunting canoe’

(Hess 1990: 174, ex. 8)

In this case, the oblique locative object subcategorized for by the bare radical becomes a direct

object of the -dxʷ form. Note, however, that in addition to transitivizing the stem, in this form the

affix also changes the semantic valency of the stem: rather than subcategorizing for a THEME (the

object aboard the canoe) and a LOCATION (the canoe), as does the bare radical √q’il, q’ildxʷ

subcategorizes for an AGENT and a LOCATION, making this use of -dxʷ quite distinct from its

normal use as a diminished control causative.

It should also be noted that the forms in (67b) and (68b) seem to lack the notion of

diminished control found in other -dxʷ forms, at least based on the glosses. It seems possible that

the uncertainty of hitting one’s target when throwing something might account for the use of the

diminished control affix with pusildxʷ, although the full-control form pusud ‘throw at

something’ exists as well. The use of the suffix in q’ildxʷ seems even more mysterious. As Hess

(1990) points out, it may be that the glosses are inadequate or in some way deceptive — or it

may simply be that these are phraseologized uses of the diminished control suffix that have gone

down their own particular path of diachronic development. Until further attestations of these

stems are uncovered, these questions will have to be left unanswered.

2.1.2.4 Causative middle -b

The suffix -b ‘causative middle [CSMD]’ is an intransitive causative suffix added to a

monovalent base to create a bivalent intransitive verb stem, increasing the valency of the radical

by adding an agentive syntactic subject, while the erstwhile PATIENT of the radical is realized as

an oblique object:

(69) a. diɬ ləsq’ʷəlas gʷəbədiɬəs

diɬ ləs–q’ʷəl=as gʷə=bə=diɬ=əs

FOC PROG.STAT–cooked=3SBJ SBJ=ADD=FOC=3SBJ

‘it’s this that would be cooked if it were that sort of thing’

(Bates, Hess & Hilbert 1994: 195)

b. huy q’ʷəlb=axʷ əlgʷəʔ ʔə tiʔəʔ buʔqʷ

huy q’ʷəl–b=axʷ əlgʷəʔ ʔə tiʔəʔ buʔqʷ

SCONJ cooked–CSMD=now PL PR PROX duck

‘well then they cook themselves these ducks’

(Hess 2006: 65, line 547)

In addition to acting as a causative and adding an AGENT to the diathesis of the verb, the

causative middle generally adds an element of self-interest on the part of the AGENT and focuses

the expression on the AGENT’s involvement in the action expressed by the verb rather than its

effect on the PATIENT. Compare the causative middle form in (69b) with the transitive (internal

causative) form in (70):

(70) bəɬčiltxʷyitəb ʔə tiʔəʔ ʔalalš ʔə tiʔəʔ sʔəɬəd ƛ’ushuys kʷi ƛ’usq’ʷəlds əlgʷəʔ

bə=ɬčil–txʷ–yi–t–əb ʔə tiʔəʔ ʔal–alš ʔə tiʔəʔ sʔəɬəd

ADD=arrive–ECS–DAT–PASS PR PROX PL–cross.sex.sibling PR PROX food

ƛ’u=s=huy=s kʷi ƛ’u=s=q’ʷəl–d=s əlgʷəʔ

HAB=NM=be.done=3PO REM HAB=NM=cooked–ICS=3PO PL

‘the brothers brought food to her again when they finished cooking it’

(Hess 2006: 45, line 72)

In this example, the focus is clearly on the fact that the food is cooked, and the AGENTs, the

brothers, do the cooking on behalf of someone else rather than in their own specific interests.

In other forms, the semantic contribution of the causative middle is less that of self-interest

and more one of placing emphasis on the AGENT’s activity, backgrounding its effect on a specific

PATIENT. This gives us verbal pairs based on the same radical such as č’aʔəb ‘dig for something

(roots)’ vs. č’aʔəd ‘dig something up’, where the middle form describes a specific kind of

digging and construes it as an activity while the internal causative describes an event and is more

focused on the specific effects (the disinterment) of an action on a particular PATIENT. Several of

these causative middle forms have lexicalized to the point of explicitly naming culturally salient

activities (e.g., ɬič’ib ‘cut something (cattails for mats)’, yiq’ib ‘make something (baskets)’).

Both aspects of the middle’s semantics — the self-interest and the activity reading — are quite in

line with the cross-linguistic behaviour of what are called middle-markers in a wide range of

languages, and fit nicely with Kemmer’s (1993) hypothesis that the middle is in general a marker

of reduced semantic transitivity.

Table 33 shows a number of forms where the middle marker is added to a monovalent radical

to form a bivalent intransitive stem:

č’aʔəb ‘dig for  (roots)’

(º√č’aʔ ‘be dug up’; cf. č’aʔəd ‘dig  up’)

gʷəč’əb ‘seek  for self’

(º√gʷəč’ ‘be sought’; cf. gʷəč’əd ‘look for ’)

kʷədab ‘take  for self’

(√kʷəd ‘be held, be taken’)

ɬič’ib ‘cut  (cattails) for mats’

(√ɬič’ ‘get cut with knife’)

ƛ’agʷəb ‘make  (mat)’

(º√ƛ’akʷ ‘be stitched’; cf. ƛ’agʷəd ‘stitch  (mat)’)

qədəb ‘have illicit sex with ’

(º√qəd ‘fornicate’; cf. dxʷqədid ‘cuckhold ’)

q’ilb ‘put  on board one’s canoe’

(√q’il ‘be aboard’)

q’ʷəlb ‘cook  for oneself’

(√q’ʷəl ‘be cooked, be ripe’)

This form is also attested as q’iləb when the middle suffix is in word-final position.

yiq’ib ‘make  (baskets)’

(√yiq’ ‘be worked into tight place’)

Table 33: Bivalent intransitive stems formed with -b

As can be seen in the forms in Table 33, the causative middle suffix has two allomorphs, [-b] and

[-əb], and interacts morphophonemically with its stem in the same way that the internal causative

suffix does (Section 2.1.2.1). For many radicals, the distribution of the two allomorphs is

conditioned by the preceding segment: the [-b] allomorph appears following vowels and

approximants (e.g., √q’ʷəl ‘be cooked, be ripe’ > q’ʷəlb ‘cook something’), and the [-əb]

allomorph appears following obstruents (√č’aʔ ‘be dug up’ > č’aʔəb ‘dig for something (roots)’).

For another set of radicals, those requiring an epenthetic harmonic vowel with the internal

causative, the causative middle also triggers final epenthesis (√ɬič’ ‘get cut with knife’ > ɬič’ib

‘cut cattails for mats’, ɬič’id ‘cut something with a knife’; √yiq’ ‘be worked into tight place’ >

yiq’ib ‘make something (baskets)’, yiq’id ‘weave something (basket)’).

Similarly, those

radicals that epenthesize a lexically-specified vowel with the internal causative add the same

vowel before the causative middle (√kʷəd ‘be held, be taken’ > kʷədab ‘take something for self’,

kʷədad ‘take something’). This morphophonemic interaction with its base distinguishes the

causative middle from the homophonous valency-neutral middle and constitutes another

argument in favour of their formal treatment as separate morphemes.

2.1.2.5 Causative of activity -alikʷ

The suffix -alikʷ ‘causative of activity [ACT]’ — or, as it has been traditionally glossed,

‘creative activity’ (Hess 1976; Bates, Hess & Hilbert 1994; Bates & Hess 2003) — is an

intransitive causative suffix which, when added to a monovalent base, creates a bivalent

intransitive verb by adding an AGENT expressed as syntactic subject. The resultant verbs express

A possible exception to this is √šab ‘be dry’ which has the middle form šabəb ‘dry something’; however, there

are no attestations of this stem that clearly illustrate its valency, so it is possible that the stem is monovalent,

meaning something like ‘dry food’. Otherwise, it may simply be an exceptional form.

an event in which the AGENT is engaged in an activity affecting a PATIENT or involving a THEME.

As an intransitive causative, however, the causative of activity creates stems that express the

PATIENT/THEME as an oblique, rather than a direct, object, as in (71):

(71) a. ʔuč’ax%ʷ čəd

ʔu–č’ax%ʷ čəd

PFV–clubbed 1SG.SUB

‘I got hit [by a branch in the thicket]’

(Bates, Hess & Hilbert 1994: 69)

b. ƛ’ubəxʷ čəɬ ʔuč’axʷalikʷ ʔə tiʔəʔ buʔqʷ

ƛ’ub=əxʷ čəɬ ʔu–č’axʷ–alikʷ ʔə tiʔəʔ buʔqʷ

well=now 1PL.SUB PFV–clubbed–ACT PR PROX duck

‘we had better use [our paddles] as clubs against these ducks’

(Hess 2006: 76, line 810)

Here, the radical č’axʷ ‘be hit with a stick’ in its bare form assigns the semantic role of PATIENT

to its single argument, which is expressed as the semantic subject (71a). When -alikʷ is added to

the radical, the subject of the derived form is an AGENT and the PATIENT is expressed as an

oblique object, as in (71b). As shown in (72), an overt, non-oblique NP is interpreted as the

subject rather than the object of an -alikʷ form:

(72) c’əlalikʷ tsiʔiɬ ƛ’aƛ’ac’apəd

c’əl–alikʷ tsiʔiɬ ƛ’aƛ’ac’apəd

defeated–ACT DIST:FEM ant

‘Ant wins’

(Hess 1995: 145, line 58)

This is an interpretive property of intransitive verbs or what have traditionally been referred to as

“agent-oriented stems” (e.g., Hess 1995).

In addition to changing the valency of its base, -alikʷ adds the notion of a repeated or

temporally extended action (Bates & Hess 2003), frequently creating verbs for culturally

important or routine activities. A number of such forms are given in Table 34:

ʔabalikʷ ‘give  away as in a potlatch’

(º√ʔab ‘be extended’; cf. ʔabəd ‘extend , give ’)

ʔilalikʷ ‘interpret ’

(√ʔil ‘sing’)

bəčalikʷ ‘bet , place a wager with ’

(√bəč ‘be lying, be fallen from standing’)

caq’alikʷ ‘spear , impale ’

(√caq’ ‘be speared, be impaled’)

cilalikʷ ‘dish  (food)’

(√cil ‘be supported, be dished up’)

cilyialikʷ ‘dish up  (food) for ’

(√cil ‘be supported, be dished up’)

c’əlalikʷ ‘defeat ’

(º√c’əl ‘be defeated’; cf. c’əld ‘defeat ’)

c’ix&alikʷ ‘fry ’

(º√c’ix& ‘be fried’; cf. c’ix&id ‘fry ’)

c’salikʷ ‘peck at ; nail ’

(º√c’əs ‘be pecked’; cf. c’əsəd ‘peck ; nail ’)

č’aʔalikʷ ‘dig for  (edible roots)’

(º√č’aʔ ‘be dug up’; cf. č’aʔəd ‘dig  up’)

č’axʷalikʷ ‘hit  with stick’

(√č’axʷ ‘be hit with a stick’)

č’ədᶻalikʷ ‘stalk  (prey)’

(º√č’ədᶻ ‘be stalked’; cf. č’ədᶻəd ‘sneak up on ’)

dᶻubalikʷ ‘dance’

(º√dᶻub ‘be kicked’; cf. dᶻubud ‘kick ’)

gəlk’alikʷ ‘knit ’

(√gəlk’ ‘be wound, be tangled’)

gʷəč’alikʷ ‘habitually seek ’

(º√gʷəč’ ‘be sought’; cf. gʷəč’əd ‘look for ’)

gʷəlalalikʷ ‘kill , slaughter ’

(º√gʷəlal ‘be hurt’; cf. gʷəlald ‘kill , punish ’)

gʷiʔalikʷ ‘ask for ; seek compensation for ’

(√gʷi ‘make an invitation’)

huyalikʷ ‘make , create ’

(√huy ‘be done, be made, be finished’)

kʷədačiʔalikʷ ‘shake hands with ’

(from √kʷəd ‘be held, be taken’ + -ačiʔ ‘hand’)

kʷədalikʷ ‘take  over and over’

(√kʷəd ‘be held, be taken’)

k’awalikʷ ‘chew ’

(º√k’aw ‘be chewed’; cf. k’awad ‘chew ’)

k’ʷɬalikʷ ‘serve  (liquid)’

(√k’ʷəɬ ‘pouring out, spill out’)

ɬač’alikʷ ‘fight fire’

(√ɬač’ ‘go out (fire)’)

ƛ’aɬəbalikʷ ‘salt ’

(√ƛ’aɬəb ‘be salty’)

p’t’alikʷ ‘save ’

(º√p’t’ ‘be stored’; cf. p’t’ad ‘store ’)

q’italikʷ ‘hang  (fish) up to dry’

(º√q’it’ ‘be hung’; cf. q’it’id ‘hang ’)

subalikʷ ‘smell ’

(º√sub ‘have odour’; cf. subud ‘smell ’)

šabalikʷ ‘dry  (food)’

(√šab ‘be dry’)

tagʷəlikʷ ‘buy ’

(√takʷ ‘be bought’)

tsalikʷ ‘hammer , pound ’

(º√ts ‘be punched’ cf. təsəd ‘punch ’)

tulalikʷ ‘read ; interpret ’

(º√tul ‘be interpreted’; cf. tulud ‘interpret ’)

tux&ʷalikʷ ‘stretch ’

(º√tux&ʷ ‘be stretched’)

t’qalikʷ ‘make bread; plaster’

(√t’q ‘be thick’)

x&ƛ’alikʷ ‘bite into ’

(º√x&ƛ’ ‘be bitten’; cf. x&əƛ’əd ‘bite ’)

xʷšalikʷ ‘sow ; give goods at potlatch’

(º√xʷš ‘be thrown’; cf. xʷəšəd ‘throw ; give away ’)

x&čalikʷ ‘count ’

(√x&əč ‘think, feel, use one’s mind’)

x&ʷadᶻalikʷ ‘slaughter ’

(º√x&ʷadᶻ ‘be injured’; cf. x&ʷadᶻad ‘punish , annihilate ’)

Table 34: Stems formed with -alikʷ

In the bulk of these forms the additional semantic component of ‘activity’ or ‘creative activity’ is

fairly obvious from the glosses, as in º√xʷəš ‘be thrown, be distributed’ vs. xʷšalikʷ ‘sow;

potlatch’ or √ɬač’ ‘go out (fire)’ vs. ɬač’alikʷ ‘fight fire’. In these forms, in addition to

causativizing the radical by adding an AGENT/subject, -alikʷ converts an expression of a state

resulting from a telic, possibly punctual, action into an expression of a non-telic activity

involving multiple instances of that action (as in the throwing of multiple seeds in sowing or the

multiple acts of distribution of goods in a potlatch), or which involve extended activity or a suite

of actions eventually leading to the resultant state (as in the various steps involved in fighting a

This form is also recorded in the Lushootseed Dictionary as šuʔbalikʷ (Bates et al. 1994: 28).

fire, all of which lead up to the fire’s extinction). The same type of distinction can be seen in

more idiosyncratic pairs such as √bəč ‘be lying’ vs. bəčalikʷ ‘bet, place a wager with’ or √gəlk’

‘be wound, be tangled’ vs. gəlk’alikʷ ‘knit’, where a fairly generic stative expression has become

lexicalized as an expression of a very specific activity involving (literally or metaphorically)

placing some PATIENT or THEME in that state. In a few cases, the lexicalized meaning is so

specific with respect to a potential object that the form is, at least in its attested uses, essentially

monovalent (e.g., ɬač’alikʷ ‘fight fire’, t’qalikʷ ‘make bread; plaster’, dᶻubalikʷ ‘dance’). Given

that the nature of the THEME of such verbs is inherently specified by the meaning of the stem, the

absence of an overtly expressed object with such forms is consistent with the focus of the

causative of activity on the ACTOR’s role in the event.

There are a few cases where the glosses of the -alikʷ form and the radical (or the transitive or

middle form of the same radical) appear to be largely synonymous, but the distinction is made

apparent by commentary from speakers or investigators on the glosses — as for, instance, with

gʷəlalalikʷ ‘kill, slaughter’ and x&ʷadᶻalikʷ ‘slaughter’ to both of which Bates, Hess & Hilbert

(1994) add the comment “for a pigʷəd (spirit power ceremony)”, or with k’ʷɬalikʷ ‘serve liquid’

to the gloss of which Bates & Hess (2003) add “repeatedly (as to all the guests)”. Similarly, to

the gloss of kʷədalikʷ ‘take over and over’ Bates & Hess (2003) add “(as in fishing)”, the

parenthesis indicating a frequent context for usage rather than a part of the semantics of the verb

(see, for instance, the use of kʷədalikʷ in Martha Lamont’s Changer story (Hess 1998: 75, line

261) where the context makes it clear that the verb there has nothing to do with catching fish).

Presumably, for the handful of apparently synonymous verbal pairs for which such commentary

is lacking, further investigation would reveal similar distinctions.

The close link between the sense of ‘activity’ inherent in -alikʷ and its frequent lexicalization

in expressions of culturally important activities is also seen in the presence of the affix in a

100

number of nominal forms expressing professions or entities defined by their behaviour, as for

example in the forms in (73):

(73) a. ƛ’uqʷatalikʷ

ƛ’u–qʷat–alikʷ

HAB–laid.out–ACT

‘farmer’

b. dukʷəlikʷ

dukʷ–əlikʷ

anormal–ACT

‘malevolent power that makes you do things the wrong way’

c. dxʷsx%ayqsalikʷ

dxʷ–s–sx%–ay–qs–alikʷ

CTD–NP–scraped–CNN–nose–ACT

‘barber’

Although these expressions are highly lexicalized, the relationship between the meaning of the

radical and the meaning of the expression as a whole is consistent with the syntax and the

semantics of the causative of activity seen in other forms such as those in Table 34.

2.1.2.6 Allative applicative -c/-s

Of the Lushootseed morphemes that fit the definition of applicative, -c/-s ‘allative applicative

[ALTV]’ is the most straightforward, the others being secondary suffixes that require the presence

of some other morpheme in order to form a transitive stem (Section 2.1.3). When attached to a

verbal base, the allative applicative adds a new argument, most frequently a GOAL, which is

realized as the direct object of the derived verb:

(74) a. huy ʔəƛ’axʷ tiʔəʔ čxʷəluʔ

huy ʔəƛ’=axʷ tiʔəʔ čxʷəluʔ

SCONJ come=now PROX whale

‘and then Whale comes’

[ML Mink and Tutyika I, line 106]

101

b. gʷəl ɬuʔəƛ’cbuɬ əlgʷəʔ čla ɬuʔa

gʷəl ɬu=ʔəƛ’–c–buɬ əlgʷəʔ čla ɬu=ʔa

then IRR=come–ALTV–1PL.OBJ PL 1PL.COORD IRR=be.there

‘then they will come for us and we will be there’

(Hess 2006: 72, line 712)

c. ɬuʔəƛ’cəb čəɬ

ɬu=ʔəƛ’–c–əb čəɬ

IRR=come–ALTV–PASS 1PL.SUB

‘we will be come after’

[ML Mink and Tutyika I, line 14]

As in (74b), the applicative object is a direct object marked by the b-series object suffixes

(Section 8.1.2) and subject to syntactic operations such as passivization (74c). When the derived

stem takes an overt NP argument, this argument is interpreted as the direct object:

(75) ʔuʔəƛ’c ti sqʷəbayʔ

ʔu–ʔəƛ’–c ti sqʷəbayʔ

PFV–come–ALTV SPEC dog

‘s/he came for the dog’

(Hess 1995: 15, ex. 10c)

Thus, allative applicative stems are ordinary transitive verbs and have all the characteristics of

transitive verbs outlined in Section 8.2.2 below.

The allative applicative morpheme has two allomorphs. The [-c] allomorph is used with a

small, idiosyncratic group of stems:

ʔəƛ’c come after ’

(√ʔəƛ’ ‘come’)

ʔəƛ’cbid ‘come after ’

(√ʔəƛ’ ‘come’)

ʔigʷəɬaac ‘climb after ’

(√ʔigʷəɬa ‘climb tree’) (Sk)

ʔux&ʷc ‘go to ’

(√ʔux&ʷ ‘go’)

baliic ‘forget about ’

(√bali ‘be forgetful’)

cuuc ‘speak to ’

(√cut ‘speak’)

čubaac ‘go inland after ’

(√čubə ‘go inland’)

day’ay’c ‘run out of ’

(√day’ ‘only’)

hədʔiw’c ‘go inside after ’

(√hədʔiw’ ‘be inside a house’)

k’ʷəƛ’c ‘miss  (target)’

(º√k’ʷəƛ’ ‘miss’; cf. k’ʷəƛ’gʷasbid ‘miss meeting’)

lax&c ‘think of ’

(√lax& ‘recall, remember’)

ləqc ‘listen to ’

(√ləq ‘listen’) (Sk)

luuc ‘listen to ’

(º√lu ‘hear’; cf. luhəladiʔ ‘hear ’)

qʷiʔaac ‘call out to ’

(√qʷiʔad ‘yell’)

q’ʷəlilc ‘warm stones to cook ’

(from √q’ʷəl ‘be cooked, be ripe’ + –ilc ‘round object’)

šuuc ‘look at ’

(√šuɬ ‘appear, be visible’)

tayc ‘come after  in raid’

(√tay ‘go raiding’)

Table 35: Stems formed with -c

102

When attached to V-final stems, the [-c] allomorph triggers lengthening of the final vowel (e.g.,

bali ‘be forgetful’ > baliic ‘forget about something’, ʔigʷəɬa ‘climb tree’ > ʔigʷəɬaac ‘climb

after something’). If the final vowel is /ə/, it becomes /aa/ (čubə ‘go inland’ > čubaac ‘go

inland after something’). For an idiosyncratic set of C-final stems, the allative applicative causes

vowel-lengthening as well as syncope of the final consonant (šuɬ ‘see’ > šuuc ‘look at’, qʷiʔad

‘yell’ > qʷiʔaac ‘call out to’).

The second allomorph, [-s], is found associated with a relatively larger group of stems, all of

which end in /il/:

ʔusis ‘dive after ’

(√ʔusil ‘dive’)

c’ip’əlis ‘ignore ’

(√c’ip’lil ‘shut eyes’)

č’itis ‘approach ’

(from č’itil ‘draw near’ based on √č’it ‘nearby’)

gʷəcis ‘wade after ’

(√gʷəcil ‘wade’)

gʷədis ‘sit down next to ’

(from gʷədil ‘sit down’ based on √gʷəd ‘down’)

həliʔis ‘live on ’

(həliʔil ‘heal’ from √həliʔ ‘be alive’)

hiwis ‘approach , go after ’

(√hiwil ‘proceed’)

lis ‘go over to ’

(√lil ‘far’)

ɬalis ‘go ashore after ’

(√ɬalil ‘go ashore’)

ɬčis ‘arrive at ’

(√ɬčil ‘arrive’)

qadils ‘come up behind ’

(qadil ‘get behind’ from √qad ‘behind’)

q’ilagʷis ‘catch a ride with ’

(q’ilagʷil ‘get aboard’ from √q’il ‘be aboard’)

qʷcagʷis ‘slide down after ’

(qʷcagʷil ‘slide down’ from º√qʷc ‘slide, slip’)

tədᶻis ‘go to bed with ’

(√tədᶻil ‘go to bed, lie in bed’)

təlawis ‘run after ’

(√təlawil ‘run’)

tudᶻis ‘bend over to get ’

(√tudᶻil ‘bend forward’)

xʷak’ʷis ‘get tired of ’

(√xʷak’ʷil ‘be tired’)

xʷt’agʷis ‘climb down after ’

(xʷt’agʷil ‘climb down’ from º√xʷit’ ‘lowered’)

x&aƛ’is ‘defend from ’

(√x&aƛ’il ‘argue’)

x&ʷubis ‘be quiet about ’

(√x&ʷubil ‘be quiet’)

Table 36: Stems formed with -s

In a few of these cases, the final sequence /il/ of the base is synchronically analyzable as either

the inchoative suffix -il or the autonomous action suffix -agʷil (which may itself be historically

analyzable as containing the inchoative suffix).

In the bulk of cases, however, the radical

without -il seems to be unattested in any environment, although generally the meaning of these

In addition to the forms found in Table 36, there is a form saxʷəbis ‘run after something’, which appears to be

based on the unattested stem *saxʷəbil (from √saxʷəb ‘jump, run), and qadis ‘approach something from behind’,

which seems to be based on *qadil (√qad ‘back up’).

103

stems is compatible with an etymological analysis that posits a historical root-plus-inchoative

combination.

Thus, diachronically, the distribution of the -s allomorph of the allative applicative

may have been due to morphological conditioning by the presence of the inchoative suffix -il,

although synchronically this seems to have been reduced to a phonological condition on the

allomorphy of the allative applicative suffix. The distribution of -il ‘inchoative’ is discussed in

Section 2.1.1.2.

2.1.3 Secondary suffixes

Secondary suffixes are affixes that combine with another suffix, usually -t ‘internal

causative’,

to form a morphological complex which increases the valency of the verb stem by

adding a direct object that expresses some semantic role other than PATIENT. In total, Hess &

Bates (2004) list four secondary suffixes — -yi-, -bi-, -di-, and -i-. Of these, only -yi- and -bi-

appear to be productive and can be associated with unique and fairly consistent meanings; the

other two are relatively infrequent and appear to be confined to a few fossilized forms and to

have meanings that overlap with those of the more productive secondary suffixes. The discussion

of these affixes here begins with -yi- (Section 2.1.3.1) and -bi- (2.1.3.2), after which the two

remaining suffixes are discussed in Section 2.1.3.3.

In fact, there is only one form in the textual corpus that contains a secondary suffix followed by a valency-

increasing morpheme other than the internal causative:

(i) gʷəl lələk’ʷəd tiʔiɬ ləcucilyialikʷ sʔəɬəd

gʷəl lə=lək’ʷ–əd tiʔiɬ ləcu–cil–yi–alikʷ sʔəɬəd

then PRG–eat–ICS DIST PROG.STAT–dish.out–DAT–ACT food

‘and as he was going along, he was eating the food that was being dished up’

(Hess 1998: 63, line 76)

The Lushootseed Dictionary also contains the forms tupyib ‘pound somethingto prepare as food’ and ƛ’al’yib ‘add

to something’. All of these involve the dative applicative, -yi-d. Hess (1967a: 43) gives the form ƛ’aldxʷid ‘manage

to get something on’, which appears to contain both the diminished control suffix -dxʷ and the secondary suffix

complex -i-d. It is not clear to what extent these verbs are fossilized forms or if they represent active derivational

possibilities in the synchronic language.

104

2.1.3.1 Dative applicative -yi-

The secondary suffix -yi- ‘dative applicative [DAT]’ combines with the internal causative

suffix -d to create trivalent transitive verbs which express an AGENT as subject and a RECIPIENT

or BENEFICIARY as direct object. When the morphological complex -yi-d is added to a

monovalent intransitive base, the effect is an increase in valency of two, as in (76):

(76) a. ʔukʷəd ti ʔiɬkʷəlq

ʔu–kʷəd ti ʔiɬ–kʷəlq

PFV–taken SPEC PRTV–other.things

‘some (not all) was taken’

(Bates, Hess & Hilbert 1994: 123)

b. ʔukʷədyic ʔə ti ɬaʔx%

ʔu–kʷəd–yi–t–s ʔə ti ɬaʔx

PFV–taken–DAT–ICS–1SG.OBJ PR SPEC platter

‘s/he took the platter from me’

(Hess 1995: 42)

c. ʔukʷədyitəb čəd ʔə tsi č’ač’as ʔə ti k’ʷat’aq

ʔu–kʷəd–yi–t–əb čəd ʔə tsi č’ač’as ʔə ti k’ʷat’aq

PFV–taken–DAT–ICS–PASS 1SG.SUB PR SPEC:FEM child PR SPEC mat

‘I had the mat taken from me by the girl’

(Hess 1995: 36, ex. 13c)

(76a) shows the monovalent radical kʷəd ‘be held, be taken’ which takes a THEME as its subject.

When -yi-d is added to the radical, the verb becomes trivalent, as in (76b). The new semantic

roles added to the radical are AGENT — the role normally added by the internal causative — and

BENEFICIARY (or perhaps in this case MALEFICIARY). Of these two new semantic actants, the

AGENT is expressed as the subject and the THEME is expressed as an oblique. The direct object is

the BENEFICIARY, which is marked using the s-series of object-markers associated with the

internal causative (Section 8.1.2). The direct object of verbs formed with -yi-d is a syntactically

ordinary direct object and is amenable to syntactic operations such as passivization (76c). An

overt, non-oblique NP appearing with a -yi-d form is interpreted as the direct object:

105

(77) ʔulək’ʷyid ti luƛ’ ʔə ti sʔuladxʷ

ʔu–lək’ʷ–yi–d ti luƛ’ ʔə ti sʔuladxʷ

PFV–eaten–DAT–ICS SPEC old PR SPEC salmon

‘s/he ate the old man’s salmon’

(Hess 1995: 36, ex. 14b)

Thus, despite being trivalent, dative applicatives form ordinary transitive clauses (Section 8.2.2).

A number of dative applicative stems formed on monovalent radicals are given in Table 37:

ʔabyid ‘give to ’

(º√ʔab ‘be extended’; cf. ʔabəd ‘extend ’)

ʔayid ‘put  there for ’

(√ʔa ‘be there’)

ʔilyid ‘sing  for ’

(√ʔil ‘sing’)

ʔux&ʷyid ‘go in place of ’

(ʔux&ʷ ‘go’)

biqʷyid ‘permit  to ’

(º√biqʷ ‘be loose’; cf. biqʷid ‘loosen ; permit ’)

cilyid ‘serve  to ’

(√cil ‘be dished up’)

hudčupyid ‘put  into the fire for ’

(from √hud ‘burn’ + -čup ‘fire’)

hudyid ‘make a fire for ’

(√hud ‘burn’)

huyid ‘make  for ’

(√huy ‘be done, be made, be finished’)

kʷədyid ‘take  from ’

(√kʷəd ‘be held, be taken’)

ləc’yid ‘step on  affecting ’

(º√ləc’ ‘have come down on’; cf. ləc’əd ‘step on ’)

lək’ʷyid ‘eat  away from ’

(√lək’ʷ ‘be eaten’)

ɬagʷidyid ‘set out a mat for ’

(from sɬagʷid ‘sleeping mat’)

ɬčilyid ‘arrive with  for ’

(√ɬčil ‘arrive’)

ɬilyid ‘give  (food) to ’

(√ɬil ‘make a gift of food’)

pqʷyid ‘break off a bit of  for ’

(º√pkʷ ‘be broken off leaving a larger piece behind’)

sulayid ‘set  before ’

(√sula ‘be in the middle of a room’)

x&ədyid ‘set  aside for ’

(º√x&əd ‘be pressed’; cf. x&ədəd ‘push ’)

x&qičyid ‘bind  into a pack for ’

(from º√x&q ‘be wrapped, be tied’ + -ič ‘bundle’)

Table 37: Stems formed with -yi-d on monovalent bases

A few of these forms have lexicalized meanings that are metaphorical or idiomatic (e.g., √sula

‘be in the middle’ > sulayid ‘set before’, º√x&əd ‘be pressed’ > x&ədyid ‘set aside for’, º√biqʷ ‘be

loose’ > biqʷyid ‘grant to, permit’). Most notable in this regard is ʔabyid ‘give to’ (from º√ʔab

‘be extended’), which is the most textually frequent of the -yi-d forms. There are also three

forms in the table which seem to be only bivalent rather than trivalent — ʔux&ʷyid ‘go in place

of’, hudyid ‘make a fire for’, and ɬagʷidyid ‘set out a mat for’. Of these, hudyid ‘make a fire

for’ and ɬagʷidyid ‘set out a mat for’ both have conventionalized THEMEs (‘wood’ and ‘mat’,

respectively) which are inherent in the semantics of the stem itself and simply may not bear

expression as an NP argument. Whether the overt use of an oblique THEME argument with these

This radical can also mean ‘be at the front of a theatre or auditorium’.

106

forms is possible or whether the absence of such forms in the corpus is merely the improbability

of an appropriate discourse context for such an argument must remain an open question for now.

The third bivalent stem, ʔux&ʷyid ‘go in place of’ is based on a monovalent agent-oriented radical

and the absence of a third syntactic argument is no doubt a consequence of the absence of a

plausible semantic role that such an argument might express. It should be noted in all three cases,

however, that the semantic role which is added by -yi-d is BENEFICIARY; this is consistent with

its behaviour in the other verb forms.

In addition to appearing with monovalent intransitive radicals, -yi-d is also found associated

with a few bivalent intransitive radicals in verbs such as such as ʔuləx&yid ‘gather something for

someone’ (√ʔuləx& ‘gather something, forage for something’), ʔəy’dxʷyid ‘find something for

someone’ (√ʔəy’dxʷ ‘find something’), haydxʷyid ‘find out about something for someone’

(√haydxʷ ‘know something’), and hiq’ʷəbyid ‘covet something of someone’s’ (√hiq’ʷəb ‘covet

something’). In these cases, although the net gain in valency is only one, the government pattern

of the resulting verb is the same as when -yi-d is added to a monovalent intransitive radical:

(78) a. ʔuʔuləx% ti luƛ’ ʔə ti bəsqʷ

ʔu–ʔuləx% ti luƛ’ ʔə ti bəsqʷ

PFV–forage SPEC old PR SPEC crab

‘the old man foraged for crab’

(Hess 1995: 28, ex. 15b)

b. ƛ’al’ čəd gʷəbəʔuləx%yid tiʔəʔ c’ix%c’ix% ʔə kʷi sʔuladxʷ

ƛ’al’ čəd gʷə=bə=ʔuləx%–yi–d tiʔəʔ c’ix%c’ix% ʔə kʷi sʔuladxʷ

also 1SG.SUB SBJ=ADD=gather–DAT–ICS PROX fish.hawk PR REM salmon

‘I too can get salmon for Fish Hawk’

(Hess 1995: 153, line 54)

In forms like these, the internal causative portion of the -yi-d complex does not causativize the

verb, but instead seems merely to indicate that the verb is transitive.

It should be noted, however, that the same pattern holds for ʔuləx& when it is combined with the internal causative

suffix alone:

107

A similar pattern is found when -yi-d is added to transitive stems formed with one of the

valency-increasing causative morphemes:

(79) a. xʷuyub

xʷuyub

be.sold

‘make a sale’

(Hess & Bates 2004: 178, ex. 14)

b. ʔuxʷuyubtubš čəxʷ

ʔu–xʷuyub–txʷ–bš čəxʷ

PFV–be.sold–ECS–1SG.OBJ 2SG.SUB

‘you sold me’

(Bates, Hess & Hilbert 1994: 255)

c. ʔuxʷuyubtxʷyid čəd tsi dʔibac

ʔu–xʷuyub–txʷ–yi–d čəd tsi d–ʔibac

PFV–be.sold–ECS–DAT–ICS 1SG.SUB SPEC:FEM 1SG.PO–grandchild

‘I sold it for my granddaughter’

(Bates, Hess & Hilbert 1994: 255)

Here, the PATIENT semantic role associated with a valency-increasing affix like the external

causative in (79b) is no longer the direct object of the verb formed with -yi-d (79c), which

expresses the BENEFICIARY in this role. When overt, the PATIENT is expressed as an oblique

object, as in (80):

(i) ƛ’uʔux%ʷ čəd čəda ƛ ’uʔuləx%əd tiʔəʔ č’itulbixʷ ʔal tiʔəʔ diʔəʔ sbadil

ƛ’u=ʔux%ʷ čəd čəda ƛ’u=ʔuləx%–əd tiʔəʔ č’itulbixʷ

HAB=go 1SG.SUB 1SG.COORD HAB=gather–ICS PROX grass

ʔal tiʔəʔ diʔəʔ sbadil

PR PROX DEM mountain

‘I go and I gather this grass on the mountain’

(Bates, Hess & Hilbert 1994: 21)

The effect of the internal causative on this particular stem is that of a simple transitivizer rather than a valency-

increasing causative morpheme. None of the other bivalent intransitive stems listed above combines with -t on its

own.

108

(80) ʔəsčal kʷi gʷədəxʷlək’ʷdxʷyids tsiʔəʔ ʔalšs ʔə tiʔəʔ sʔəɬəds

ʔəs–čal kʷi gʷə=dəxʷ=lək’ʷ–dxʷ–yi–d=s tsiʔəʔ

STAT–how REM SBJ=ADNM=eaten–ICS–DC–DAT–ICS=3PO PROX:FEM

ʔalš–s ʔə tiʔəʔ sʔəɬəd–s

cross.sex.sibling–3PO PR PROX food–3PO

‘how could he eat his sister’s food away from her?’

(Hess 1998: 56, line 6)

Here, the oblique object of the verb lək’ʷdxʷyid ‘eat something away from someone, manage to

get someone’s food and eat it’, tiʔəʔ sʔəɬəds ‘her food’, corresponds to the PATIENT/direct object

of the plain transitive form lək’ʷdxʷ ‘mange to eat something’. Once again, the internal causative

portion of the -yi-d complex seems not to function so much as a valency-increasing affix as it

does as a marker of the transitivity of the clause

As these examples show, while the basic syntactic effect of -yi-d is to increase the valency of

a verb stem, it may not increase it beyond the upper limit of three syntactic arguments. If the

verb stem is monovalent intransitive, its valency is increased by two, as in (76); if the form is

bivalent intransitive, its valency is increased by one and the stem is transitivized, as in (78); if the

stem is already transitive, the valency is increased by one and the government pattern is altered

so that what was expressed as the direct object of the transitive form becomes an oblique object

of the -yi-d form, as in (80). Although the effect on the stems is different, the government pattern

of the resulting verb is always the same — a trivalent transitive verb with a

RECIPIENT/BENEFICIARY expressed as direct object and a PATIENT/THEME expressed as an oblique.

A number of bivalent stems that take -yi-d are given in Table 38:

ʔaladᶻiʔlyid ‘babysit  for ’

(from º√ʔaladᶻ ‘care for ’ + -iʔɬ ‘child’)

ʔəy’dxʷyid ‘find  for ’

(√ʔəy’dxʷ ‘find ’)

ʔuləx&yid ‘gather  for ’

(√ʔuləx& ‘gather , forage for ’)

cildxʷyid ‘serve  to ’

(√cil ‘be dished up’ + -dxʷ ‘diminished control’)

haydxʷyid ‘find out about  for ’

(º√hay ‘be known’; cf. haydxʷ ‘know ’)

hiq’ʷəbyid ‘covet  from ’

(√hiq’ʷəb ‘covet ,lust after ’)

huydxʷyid ‘set up  for ’

(√huy ‘be done’ + -dxʷ ‘diminished control’)

kʷədabyid ‘make  captive’

(√kʷəd ‘be held, be taken’ + -b ‘causative middle’)

kʷukʷcutyid ‘cook  for ’

(√kʷukʷcut ‘cook ’)

lək’ʷdxʷyid ‘manage to eat  away from ’

(√lək’ʷ ‘be eaten’ + -dxʷ ‘diminished control’)

ɬčiltxʷyid ‘bring  for ’

(√ɬčil ‘arrive’ + -txʷ ‘external causative’)

ɬəgʷlyid ‘leave  for ’

(√ɬəgʷɬ ‘leave ’)

109

pusilyid ‘throw  for ’

(√pus ‘be hit by  (missile)’ + -il ‘inchoative’)

qadadyid ‘steal  for ’

(√qada ‘steal ’ + -t ‘internal causative’)

qʷuʔqʷadyid ‘drink  (drink) of ’s’

(√qʷuʔqʷa ‘have a drink’ + -t ‘internal causative’)’

təxʷtxʷyid ‘buy  for  ‘

(√təxʷ ‘buy ’ + -txʷ ‘external causative’)

xʷuyubtxyidʷ ‘sell  for ’)

(º√xʷuyub ‘be sold’ + -txʷ ‘external causative’)

Table 38: Stems formed with -yi-d on bivalent bases

The forms in this table are built on both bivalent intransitive and bivalent transitive bases. The

bulk of the transitive bases contain one of the causative valency-increasing affixes (-txʷ ‘external

causative’, -dxʷ ‘diminished control’, or -b ‘causative middle’), although there are two

inherently transitive forms — √ʔəy’dxʷ ‘find something’ and √ɬəgʷɬ ‘leave something’. The first

of these is clearly diachronically derived from a -dxʷ form, while the third, √ɬəgʷɬ, is one of the

few inherently transitive radicals with no identifiable derivational history. In addition, there is

pusil ‘throw something’ which is formed from the radical √pus ‘be hit by something (missile)’

and an idiosyncratic implement use of the inchoative suffix -il (Section 2.1.1.2). The remainder

of the -yi-d forms in Table 38 are based on bivalent intransitive radicals. It is also worth noting

that the forms qadadyid ‘steal something for someone’ and qʷuʔqʷadyid ‘drink something of

someone’s’ appear to be based on unattested internal causative stems *qadad ‘steal something’

and *qʷuʔqʷad ‘drink something’; in the case of qadadyid, there is an attested bivalent

intransitive form √qada ‘steal something’.

2.1.3.2 Middle applicative -bi-

The secondary suffix -bi- ‘middle applicative [MAP]’ combines with the internal causative

suffix -t to form transitive stems whose direct object expresses semantic actants in a variety of

roles other than PATIENT. The range of semantic roles, and to a certain extent the syntactic effect

of -bi-d on its stem, is much more variable than it is for -yi-d, and the over-arching semantic

linkage amongst the different uses of this morphological complex seems to be a rather abstract

notion of reduced semantic transitivity (Hopper & Thompson 1980), a notion identified by

Kemmer (1993) as being the common thread linking middle forms across a wide range of

110

languages. The parallel is strengthened by the overlap in semantic domain with the valency-

neutral middle -b (Section 2.1.1.3) in Lushootseed and the cognate -m and -mi suffixes in other

Salishan languages, elements which also cluster in their meanings around the prototypical

meanings of the middle.

The feature of -bi-d that distinguishes it most clearly from the ordinary middle suffix is its

syntactic effects on the stem to which it is attached. Because middles are generally associated

with reduced semantic transitivity, their most common syntactic effect cross-linguistically is to

detransitivize —or intransitivize — a stem; -bi-d, however, most often has the opposite effect on

syntactic transitivity and increases the valency of a stem by adding a direct object:

(81) a. ʔuʔup’ čəd

ʔu–ʔup’ čəd

PFV–be.seated.on.lap 1SG.SUB

‘I sat on a lap’

(Bates, Hess & Hilbert 1994: 22)

b. ʔəsʔup’bid čəd tiʔiɬ

ʔəs–ʔup’–bi–d čəd tiʔiɬ

STAT–be.seated.on.lap–MAP–ICS 1SG.SUB DIST

‘I’m sitting on his lap’

(based on Bates, Hess & Hilbert 1994: 22)

When the object is first- or second person, it is expressed by the s-series of object-markers:

(82) ƛ’ub čəxʷ ʔušəbic čxʷa baɬac

ƛ’ub čəxʷ ʔušəb–bi–t–s čxʷa baɬa–t–s

well 2SG.SUB pity–MAP–ICS–1SG.OBJ 2SG.COORD cure–ICS–1SG.OBJ

‘you should take pity on me and perform a shaman cure for me’

(Hess 1998: 57, line 32)

Like all direct objects, the argument added to the verb stem by -bi-d can be promoted by

passivization to become a subject:

The verb form is given in the source as ʔəsp’up’bid, although the sub-entry heading is the expected form ʔup’bid,

as is the verb form in the subsequent example. The form ʔup’bid is also found in Hess & Bates (2004: 180, ex. 23).

111

(83) a. hay, x%ʷul’əxʷ əlgʷəʔ ʔəshiq’ʷabid tiʔəʔ qa ʔəsqʷat

hay x%ʷul’=əxʷ əlgʷəʔ ʔəs–hiq’ʷab–bi–d tiʔəʔ qa ʔəs–qʷat

INTJ only=now PL STAT–covet–MAP–ICS PROX much STAT–laid.out

‘well then they just coveted the many (dentalia) that were lying there’

(Hess 2006: 60, line 439)

b. gʷəl ʔəshiq’ʷabitəb ʔə tiʔəʔ sbiaw tsiʔəʔ čəgʷas sx%aʔhus

gʷəl ʔəs–hiq’ʷab–bi–t–əb ʔə tiʔəʔ sbiaw tsiʔəʔ čəgʷas sx%aʔhus

then STAT–covet–MAP–ICS–PASS PR PROX coyote PROX:FEM wife sawbill

‘then this wife, Sawbill, was coveted by Coyote’

(Hess 2006: 22, line 12)

Non-oblique NP arguments with -bi-d stems are interpreted as direct objects:

(84) x%əcbidəxʷ tsiʔəʔ čəgʷas

x%əc–bi–d=əxʷ tsiʔəʔ čəgʷas–s

afraid–MAP–ICS=now PROX:FEM wife–3PO

‘he is afraid of his wife now’

(Hess 2006: 6, line 78)

Thus, like all applicative objects, the object of stems formed with -bi-d is a morphosyntactically

ordinary direct object.

The most consistent pattern found with -bi-d derivations is one where a monovalent

intransitive base is transitivized by the addition of a second semantic actant, in Table 39:

ʔəɬdiluɬbid ‘go to eat off of ’

(from ʔəɬdiluɬ ‘go out to eat’)

ʔup’bid ‘sit on ’s lap’

(√ʔup’ ‘be seated on a lap’)

gʷahbid ‘accompany ’

(√gʷah ‘accompany, go along’)

laqbid ‘be behind ’

(√laq ‘be last’)

lax&bid ‘remember ’s story’

(√lax& ‘recall, remember’)

ɬaw’tbid ‘be new for ’

(√ɬaw’t ‘be new’)

saxʷəbid ‘run after  or up to ’

(√saxʷəb ‘jump, sprint’)

sulabid ‘be in the middle of a room relative to ’

(√sula ‘be in the middle of a room’)

šuɬbid ‘expect , keep an eye out for ’s arrival’

(√šuɬ ‘appear, be visible’)

təlčbid ‘miss  (throwing)’

(√təlč ‘be wide of mark’)

t’q’abid ‘put stickum on ’

(√t’q’ ‘be patched (with sticky substance)’)

xʷak’ʷilbid ‘become disaffected with ’

(√xʷak’ʷil ‘be tired’)

x&əčbid ‘intend ’

(√x&əč ‘think, feel, use one’s mind’)

x&ʷal’bid ‘be unable to manage ’ ,

(√x&ʷal’ ‘be unable, fail, lose’)

wačbid ‘watch ’

(√wač ‘keep watch’)

yayusbid ‘work on ’

(√yayus ‘do work’)

yəyəhubid ‘tell  a traditional story’

(from yəyəhub ‘tell a traditional story’)

Table 39: Applicative uses of -bi-d

As noted earlier, the specific semantic roles played by the new actants vary quite a bit from verb

to verb. In several cases, the new role seems to be locative (e.g., ʔup’bid ‘sit on someone’s lap’,

112

sulabid ‘be in the middle of a room relative to something’) or directional (saxʷəbid ‘run after

something or up to something’, wačbid ‘watch something’), whereas in others -bi-d seems

simply to be adding whatever kind of additional role might plausibly be associated with a

particular type of event. An interesting contrast is found between the forms laxdxʷ ‘remember

something’ and lax&bid ‘remember someone’s story, remember the whole situation regarding

someone’, where the difference seems to be one of thinking specifically about a person versus

recalling not so much that person directly as a set of events surrounding the individual. The

common thread running through all of these forms is that the -bi-d object is not directly affected

by the action in the way that PATIENT would be — in other words, the interaction between the

AGENT and the UNDERGOER/ENDPOINT of the event is less semantically transitive than the typical

interaction between an AGENT and a PATIENT, where the PATIENT undergoes some internal change

of state.

To a certain extent the less-direct interaction between AGENT and ENDPOINT found with the

middle applicative parallels in some ways the less-direct interaction between the AGENT and the

GOAL in allative applicative constructions. Two of the radicals in Table 39 have both a -bi-d and

a -c form. One of these is √šuɬ ‘appear, be visible’, which is the base for šuɬbid ‘keep an eye out

for someone’s arrival’ and suuc ‘look at something’. The contrast in the semantic roles of the

objects in these two forms is fairly clear: in the allative applicative form the EXPERIENCER’s gaze

is directed towards an object which is present and which can serve as a specific locus on which

his/her attention is focused (i.e., a metaphorical GOAL for one’s attention), while in the middle

applicative form the potential PERCEPT is not present and the EXPERIENCER is not (yet) interacting

with it. So the distinction here is both one of difference in semantic role (GOAL vs. non-GOAL)

and in semantic transitivity, the -bi-d form being much lower on that particular scale. The second

stem, xʷak’ʷilbid ‘become disaffected with something, tire of something due to one’s own lack

David Beck 10-2-7 2:19 PM

Comment: note the odd effect on the

valency, given the unaccusative radical

113

of enthusiasm or energy’, contrasts with xʷak’ʷis ‘become fed up with something tiresome’,

formed from the radical √xʷak’ʷil ‘be tired’ and the allative applicative. The distinction here

seems to be a distinction in the locus of the impetus of the event: in the -bi-d form the source of

the feeling of disaffection is internal, whereas in the allative form the impetus comes from the

nature of the STIMULUS. While this is not easily characterized in terms of a distinction in

semantic transitivity, it is consistent with the characterization of -bi-d as a middle, given that

middles cross-linguistically are associated with the interests of the AGENT/EXPERIENCER/subject

and are frequently used to express events in which the impetus for an event is internal to the

ACTOR.

Reduced semantic transitivity in the form of the lack of direct-affectedness of the object by

the subject in -bi-d constructions is seen quite clearly in a rather large group of stems in which

the semantic role of the applicative object is that of MOTIVE:

ʔukʷukʷbid ‘make fun of ’

(√ʔukʷukʷ ‘play, have fun’)

ʔušəbid ‘feel pity for ’

(º√ʔušəb ‘feel pity’)

c’adᶻax&bid ‘be bothered by , be fed up with ’

(º√c’adᶻax& ‘feel annoyance’; cf. c’adᶻaxtx&ʷ ‘bother ’)

c’ip’lilbid ‘shut eyes to avoid seeing ’

(√c’ip’lil ‘shut eyes’)

dxʷcutəbid ‘catch on to ’

(dxʷcutəb ‘think something’ from √cut ‘speak’)

dᶻaƛ’bid ‘be confused by , forget ’

(√dᶻaƛ’ ‘be confused’)

dᶻaqəbid ‘mourn for ’

(º√dᶻaq ‘be in mourning’; cf. dᶻaqad ‘mourn ’)

hiiɬbid ‘be happy about ’

(√hiiɬ ‘be happy’)

ǰuʔilbid ‘be happy for ’

(from º√ǰuʔ ‘be glad’ + -il ‘inchoative’)

pitəbid ‘pay attention to ’

(√pitəb ‘pay attention, be aware’)

q’albid ‘be fooled by ’

(º√q’al ‘be fooled, be convinced’; cf. q’alad ‘fool ’

x&ayəbid ‘laugh at ’

(√x&ayəb ‘laugh’)

x&əcbid ‘fear ’

(√x&əc ‘be afraid’)

x&ix&ibid ‘be ashamed of ’

(√x&ix&iʔ ‘be ashamed’)

x&ʷaq’ʷbid ‘be concerned about ’

(√x&ʷaq’ʷ ‘be worried, be preoccupied’)

yabuk’ʷbid ‘fight over ’

(√yabuk’ʷ ‘(to) fight’)

Table 40: Applicative stems formed with -bi-d expressing motive

The majority of the verbs in Table 40 are based on radicals expressing mental states or emotions,

the applicative object being the STIMULUS or MOTIVE for the experience. Two of the verbs —

c’ip’lilbid ‘shut eyes to avoid seeing something’ and yabuk’ʷbid ‘fight over something’ — are

based on radicals expressing more concrete actions; in both cases, the applicative objects are

114

clearly MOTIVEs for the event. In no case is the actant expressed by the object of any of these

verbs necessarily affected by the actions performed or the emotions experienced by the ACTOR.

A third set of -bi-d stems is formed in combination with lexical suffixes. Several of these are

given in Table 41:

c’ic’əyikʷalusbid ‘wink at ’

(from º√c’ic’əyikʷ ‘wink’ + -alus ‘eye’)

dᶻalqʷusbid ‘look over shoulder at ’

(from √dᶻal ‘present other side’ + -us ‘face’)

dᶻəlax&adbid ‘visit ’

(from √dᶻal ‘present other side’ + -ax&ad ‘side’)

dᶻəlulčbid ‘turn towards ’

(from √dᶻal ‘present other side’ + -ulč ‘belly’)

ləqaladiʔbid ‘overhear ’

(Sk)

(from √ləq ‘listen’ + -al-adiʔ ‘ear’)

ɬaʔačiʔbid ‘touch  with hand’

(from √ɬaʔ ‘arrive at a specific place’ + -ačiʔ ‘hand’)

ɬadəyʔlucidbid ‘address  as woman’

(from √ɬadəyʔ ‘woman’ + -l-ucid ‘mouth’)

tubšlucidbid ‘address  as man’

(from √tubš ‘man’ + -l-ucid ‘mouth’)

xʷəbaličbid ‘toss  (pack) onto own back’

(from √xʷəb ‘toss’ + -alič ‘bundle’)

x&ʷil’alcbid ‘lose ’

(from √x&ʷil’ ‘be lost’ + -alc ‘object’)

yəlačiʔbid ‘use both hands on ’

(from º√yəl ‘pair’ + -ačiʔ ‘hand’)

Table 41: Stems formed with -bi-d and lexical suffixes

As with the verbs in the earlier sets, the stems here take a non-PATIENT object — one which is

not directly affected by the action of the AGENT by undergoing an internal change of state — and

the specific roles played by the objects are rather diverse. These range from PERCEPT (laʔačiʔbid

‘touch something with hand’, ləqaladiʔbid ‘overhear something’) to DIRECTION/GOAL (c’ic’əyi-

kʷalusbid ‘wink at something’, dᶻəlax&adbid ‘visit someone’), HEARER (ɬadəyʔlucidbid ‘address

someone as woman’), or various types of THEME (laʔačiʔbid ‘touch something with hand’, xʷəb-

aličbid ‘toss something (pack) onto own back’, x&ʷil’alcbid ‘lose something’). Hess & Bates

(2004) point out that in these constructions the lexical suffix expresses a bodypart playing an

instrument-like role in the event. Verbs expressing action directed towards or involving parts of

an AGENT’s body are commonly middle forms across languages (Kemmer 1993), and the

Note that, due to a typographical error, the Lushootseed Dictionary glosses the stem c’ip’lis ‘shut eyes to avoid

getting something in them’, formed with the allative applicative, as having a second reading synonymous with

c’ip’lilbid ‘shut eyes to avoid seeing something’. In fact, c’ip’lis means only ‘shut eyes to avoid getting something in

them’.

Also ləqəladiʔbid.

Also sɬadəyʔlucidbid.

115

lowered affectedness of the object (and, hence, the reduced semantic transitivity of the event) is

typical of middle semantics.

There is at least one verb form in which -bi-d, like -yi-d, seems to increase the valency of its

base by two rather than by one — saxʷəbid ‘run away with something of someone’s’ (see the

homophonous form saxʷəbid ‘run after something or up to something’ in Table 39):

(85) diɬəxʷ sx%aab ʔə tsiʔəʔ sɬadəyʔ ʔusaxʷəbitəb ʔə tiʔəʔ tubədaʔs

diɬ=əxʷ s=x%aab ʔə tsiʔəʔ sɬadəyʔ ʔu–saxʷəb–bi–t–əb

FOC=now NM=cry PR PROX:FEM woman PFV–run–MAP–PASS

ʔə tiʔəʔ tu=bədaʔ–s

PR PROX PAST=offspring–3po

‘it is thus that the woman whose child was run away with is crying’

[HM Star Child, line 59]

Here, the verb form in question is in the passive, contained inside a subject-centred relative

clause modifying sɬadəyʔ ‘woman’. The subject of the passive is the MALEFICIARY,

corresponding to the direct object of the active form, while the THEME, tiʔəʔ tubədaʔs ‘her child’

(lit. ‘her former/ex-child’) is realized as an oblique object, following the regular pattern for

derivations with -yi-d. Indeed, given the semantic role assigned to the object, we might have

expected the form to be *saxʷəbyid; however, this form is unattested.

Also like -yi-d, -bi-d combines with a small number of bivalent bases whose valency remains

unchanged:

hiq’ʷəbid ‘lust after ’

(√hiq’ʷəb ‘lust after ’)

kʷədabid ‘take  captive’

(kʷədab ‘capture ’ from √kʷəd ‘be taken’)

qadabid ‘steal ’

(√qadaʔ ‘steal ’)

q’ʷuʔbid ‘be together with ’

(√q’ʷuʔ ‘be together with ’)

Table 42: Valency-neutral uses of -bi-d

In these cases (hiq’ʷəbid ‘lust after someone’, kʷədabid ‘take someone captive’, qadabid ‘steal

something’, q’ʷuʔbid ‘be together with someone’), -bi-d combines with a bivalent intransitive

stem to create a transitive verb, and so acts merely as a syntactic transitivizer. The glosses given

for the -bi-d forms and their bases are significantly different in only two cases — kʷədabid ‘take

116

someone captive’ and ɬəgʷəlbid ‘leave someone behind, leave someone’s presence’. In the

remainder of the examples, the two forms seem to be nearly synonymous; however, in most of

these, the -bi-d stems take objects that are human — that is ‘someone’ rather than ‘something’.

The bulk of these verbs express actions (e.g., ‘lust after someone’, ‘be together with someone’)

that most naturally have human endpoints, and even those that do not seem to require a human

object fro semantic reasons, such as qadabid ‘steal something’, have these exclusively in their

textual attestations:

(86) huy, yəcəbaxʷ ʔə tiʔəʔ sqadabitəbs ʔə tsiʔəʔ sxʷəyuq’ʷ tiʔiɬ wiw’su

huy yəc–əb=axʷ ʔə tiʔəʔ s=qada–bi–t–əb=s ʔə tsiʔəʔ

SCONJ reported–MD=now PR PROX NM=steal–SS–ICS–PASS=3PO PR PROX

sxʷəyuq’ʷ tiʔiɬ wiw’su

Basket.Ogress DIST children

‘he told about the stealing of the children by the Basket Ogress’

[ML Basket Ogress, line 54]

In this sentence, the human NP tiʔiɬ wiw’su ‘those children’ is the subject of the passive form of

the verb, thereby corresponding to the direct object of the active form. Although the number of

textual attestations of all of the -bi-d forms in Table 42 is limited, they all seem to involve

human objects and the semantic roles played by the objects of these forms is consistent with

other uses of -bi-d: they are not canonically PATIENT-like in that the semantic ENDPOINTs of the

events do not undergo any internal change of state. Nevertheless, it should be noted that the

bivalent bases for these -bi-d forms have the same glosses, and in at least some cases their

objects can also be human:

(87) a. xʷiʔ kʷ adsʔəƛ’cbuɬ

xʷiʔ kʷi ad=s=ʔəƛ’–c–buɬ

NEG REM 2SG.PO=NM=come–ALTV–1PL.OBJ

‘don’t come to us’

(Hess 2006: 32, line 263)

David Beck 10-2-7 2:19 PM

Comment: don’t forget about this in the

discussion of the deictics

117

b. … tsiʔəʔ bədaʔs səshiq’ʷəbs

tsiʔəʔ bədaʔ–s s=ʔəs–hiq’ʷəb=s

PROX:FEM offspring–3PO NM=STAT–lust.for=3PO

‘… his daughter after whom he lusted’

(Hess 1998: 95, line 131)

Thus, it seems that teasing out whatever semantic distinctions there are between the pairs of verb

forms in Table 42 will depend on uncovering further textual attestations; nevertheless, the middle

applicative forms themselves — transitive verbs with a non-PATIENT object — are entirely

typical of -bi-d derivations.

In a few other stems, -bi-d seems to act as a causative, adding an AGENT/subject to a

monovalent radical rather than adding an object:

ʔadᶻqbid ‘meet ’

(º√ʔadᶻq ‘be met’; cf. ʔadᶻqdxʷ ‘happen to meet ’)

čəgʷasbid ‘take  as wife’

(√čəgʷas ‘wife’)

kʷədbid ‘steal from ’

(√kʷəd ‘be held, be taken’)

p’ayəqbid ‘hew , carve ’

(√p’ayəq ‘carve canoe’)

qəlbidbid ‘discard ’

(qəlbid ‘garbage’ from √qəl ‘bad’ + -bid ‘instrument’)

suxʷtəbid ‘recognize ’

(º√suxʷt ‘be recognized’; cf. suxʷtəš ‘recognize ’)

yəcbid ‘tell about ’

(º√yəc ‘be reported’; cf. yəcəd ‘report ’)

Table 43: Causativizing uses of -bi-d

Although the syntactic effect of -bi-d on its base in these forms resembles the effects of a

causative morpheme such as -t, -txʷ, or -dxʷ, the true causatives create verbs that express events

in which an AGENT acts upon a PATIENT or some other semantic actant in a PATIENT-like semantic

role. With the exception of p’ayəqbid ‘hew something, carve something’,

the objects of the -bi-

d forms in Table 43 are non-PATIENTs and do not undergo an internal change of state as a result

of the AGENT’s actions: instead, the change experienced by the object of such verbs seems to

reside more generally in its relationship to the AGENT (čəgʷasbid ‘take someone as wife’,

suxʷtəbid ‘recognize someone’, qəlbidbid ‘discard something’ [lit. ‘cause something to be

refuse to one’]) or as a point of reference — literal (ʔadᶻqbid ‘meet someone’) or figurative

The exceptional nature of p’ayəqbid ‘hew something, carve something’ may stem from the involvement of one’s

spirit power in the carving of a canoe, -bi- indicating a reduced semantic transitivity that comes either from the idea

that the AGENT is acting indirectly through an intermediary, or that the primary interaction is between the carver and

the spirit-power, and the product of the interaction is construed as less directly involved than a prototypical PATIENT.

118

(kʷədbid ‘steal from someone’) — for the AGENT’s action. The reduced semantic transitivity of

such forms is clear. The fact that -bi-d adds an AGENT/subject in these forms rather than a non-

PATIENT object, as it does more regularly, may have to do with the nature of the radicals, which

are (with the exception of čəgʷasbid ‘take someone as wife’ and qəlbidbid ‘discard something’,

which are based on nouns) “patient-oriented” in the sense that they express states that are the

outcome of events involving the interaction of two or more participants, and express the

ENDPOINT of the event as their syntactic subject. However, given the relative scarcity of such

forms, it seems likely that this is only a diachronic or a post-hoc explanation, and the forms in

Table 43 will have to be treated as lexicalized uses of what is otherwise an applicative

morpheme.

The secondary suffix sequence -bi-d also appears in at least four forms following another

valency-increasing affix — specifically, the allative applicative. These forms are given in Table

44:

lčisbid ‘visit and inconvenience them’

(from ɬčis ‘arrive at ’ based on √ɬčil ‘arrive’)

šuucbid ‘keep an eye out for ’

(from šuuc ‘look at ’ based on √šuɬ ‘appear, be visible’)

tədᶻisbid ‘go to bed with  for sex’

(from tədᶻis ‘go to bed with ’ based on √tədᶻil ‘go to bed’)

xʷakʷisbid ‘tire of  (person)’

(from xʷak’ʷis ‘get tired of ’ based on √xʷak’ʷil ‘be tired’)

Table 44: Stems formed with -bi-d and the allative applicative

In these stems, the affixation of the middle applicative has no effect on the syntactic valency of

its base, nor does it have any great effect on the semantic role of the applicative object. Its major

effect is to modify the event expressed by the allative stems in more subtle ways. In one of these

cases, xʷakʷisbid ‘tire of someone’, -bi-d seems to indicate that the applicative object is animate

or human (cf. some of the valency-neutral forms in Table 42 above). This may also be the case

for šuucbid ‘keep an eye out for someone’, although there are no contextualized examples

currently available in the textual corpus or in published sources. The remaining two verbs,

lčisbid ‘visit someoneand inconvenience them’ and tədᶻisbid ‘go to bed with someone for sex’

also necessarily have human objects, but differ in other — rather idiosyncratic ways — from

119

their allative forms. Clearly, judged by the unusual syntactic and semantic effects of -bi-d in

these forms, the stems in Table 44 are lexicalized forms and, although they are not entirely out of

step with more transparent middle applicatives, they can not be treated as synchronically

compositional forms.

Not unexpectedly, there are a number of stems that appear to contain -bi-d but are not

synchronically transparent or analyzable. Two of these are q’itbid ‘store something (food)’

(apparently based on an otherwise unattested radical *√q’it ‘be stored’), pək’ʷibid ‘snatch

something’ (based on *√pək’ʷib ‘snatch something’), and yičəbid ‘observe something’ (based on

*√yičəb ‘observe’). Among the idiomatic forms are qʷic’bid ‘be unable to do something’ (from

√qʷic’ ‘be indifferent, be lazy about’), qʷuʔbid ‘mouth waters for something’ (from the nominal

radical √qʷuʔ ‘water’), and x&ʷil’alcbid ‘lose something’ (√x&ʷil’ ‘be lost’ + -alc ‘object’).

2.1.3.3 Other secondary suffixes -di-, -i-

In addition to -yi-d and -bi-d, there are two more secondary suffix complexes that act as

applicatives. One of these, -di-d, appears as part of four stems, given in Table 45:

dxʷqədid ‘have sex with  (spouse)of ’

(from dxʷ- ‘contained’ + º√qəd ‘fornicate’)

punishdid ‘punish ’

(from Eng. punish)

qadadid ‘steal  from ’

(√qadaʔ ‘steal ’)

qʷuʔqʷadid ‘drink ’

(√qʷuʔqʷa ‘have a drink’)

Table 45: Stems formed with -di-d

Even in this small set of verbs, there is a great deal of variation in the effects of -di-d on the

valency and government pattern of the base to which it is attached. In two of the cases, qadadid

‘steal something from someone’ and qʷuʔqʷadid ‘drink something’, the secondary suffix

complex has a clearly applicative effect, adding a direct object to the clause. As noted by Hess &

Bates (2004), however, the objects of -di-d forms do not consistently express a particular

semantic role. The object of qʷuʔqʷadid ‘drink something’ is clearly a PATIENT, at least to the

extent that a liquid undergoes an internal change of state when it is drunk (otherwise, it is a

120

THEME), whereas the direct object of qadadid ‘steal something from someone’ is a

MALEFICIARY. Hess & Bates (2004) point out that the latter form co-exists with a -yi-d stem

based on the same radical:

(88) a. ʔuqadaditəb čəd ʔə ti dsduukʷ

ʔu–qada–di–t–əb čəd ʔə ti d–sduukʷ

PFV–steal–SS–ICS–PASS 1SG.SUB PR SPEC 1SG.PO–knife

‘I had my knife stolen’

(Bates, Hess & Hilbert 1994: 172)

b. ʔuqadadyitəb ti luƛ’

ʔu–qada–d–yi–t–əb ti luƛ’

PFV–steal–ICS–DAT–PASS SPEC old

‘the old man was stolen from’

(Bates, Hess & Hilbert 1994: 173)

The same verb form in (88b) also appears in examples glossed ‘steal for someone’, whereas the

-di-d form in (88a) has only the gloss ‘steal from someone’, leading to the conclusion that -yi-d

is more closely associated with the BENEFICIARY/MALEFICIARY semantic role whereas -di-d may

be (like -bi-d) more an indicator of a less-specific non-PATIENT role which is interpreted as

MALEFICIARY because of the nature of the event (an implicit third semantic role in a theft being a

victim). However, it should be noted that the form qadadyid appears to be based on an

unattested stem *qadad, and so has a more complicated derivational history than most -yi-d

forms. Thus, the co-existence of qadadid and qadadyid may have more to do with historical

developments in the language than with a consistent semantic contrast between the two

secondary suffixes.

The remaining two forms are even less helpful in sorting out the meaning and syntactic

behaviour of -di-d. In the case of punishdid ‘punish someone’, the precise effects of -di-d on its

base are hard to pinpoint because the “radical” punish is an English borrowing that would not

normally be used on its own in the language and so is of indeterminate valency (for Lushootseed

speakers). The verb dxʷqədid ‘have sex with someone’s spouse’ is trivalent rather than

121

monovalent, as shown in (89a), and is based on a radical that otherwise only appears with the

middle suffix -b, as in (89b):

(89) a. diɬ əw’ə higʷəxʷ ʔudxʷqədidəxʷ tiʔəʔ sʔušəbabdxʷ sbəq’ʷaʔ ʔə tsiʔəʔ čəgʷas

x%ʷuʔx%ʷəyʔ

diɬ əw’ə higʷ=əxʷ ʔu–dxʷ–qəd–di–d=əxʷ tiʔəʔ sʔušəbabdxʷ

FOC PTCL big=now PFV–CTD–fornicate–SS–ICS=now PROX unfortunate

sbəq’ʷaʔ ʔə tsiʔəʔ čəgʷas–s x%ʷuʔx%ʷəyʔ

heron PR PROX:FEM wife–3PO helldiver

‘indeed, he is the one who really cuckolded poor Heron with his wife, Helldiver’

(Hess 2006: 14, line 77)

b. ʔuqədəb əw’ə ʔə tiʔəʔ cədiɬ sč’ətx% tiʔəʔ tushuy ʔə tsiʔəʔ x%ʷuʔx%ʷəyʔ

ʔu–qəd–əb əw’ə ʔə tiʔəʔ cədiɬ sč’ətx% tiʔəʔ tu=s=huy

PFV–fornicate–MD PTCL PR PROX he kingfisher PROX PAST=NM=be.done

ʔə tsiʔəʔ x%ʷuʔx%ʷəyʔ

PR PROX:FEM helldiver

‘what Helldiver did [was] have sex with Kingfisher’

(Hess 2006: 21, line 238)

Once again, attributing a specific effect of -di-d on its base in this form is rather difficult as the

radical otherwise is attested only as part of a middle form — possibly a causative middle, given

that the verb qədəb is bivalent (see Section 2.1.1.3 below). This might lead us to surmise that the

radical itself is more amenable to a patient-oriented gloss rather than the agent-oriented gloss

‘fornicate’. This would make -di-d more like -yi-d both semantically and syntactically, as it both

causativizes the radical — adding an AGENT — and acts as an applicative, adding a

MALEFICIARY. Nevertheless, with only four forms containing -di-d, little more can be said of it

than that it is a historical relic, probably a remnant of an older secondary suffix used in the

formation of some kind of applicative, and whose effects on a particular base are today

essentially idiosyncratic.

Hess & Bates (2004) also point to a fourth secondary suffix complex, -i-d, which is most

robustly associated with lexical suffixes. A number of such forms is given in Table 46:

ʔabucidid ‘bring  lunch’

(from º√ʔab ‘be extended’ + -ucid ‘mouth’)

č’əlpačiʔid ‘twist ’s wrist’

(from √č’əlp ‘sprain, turn’ + ačiʔ ‘hand’)

122

dxʷcaq’ax&adid ‘spear in the side’

(from √caq’ ‘be speared’ + -ax&ad ‘side’)

dxʷpuhigʷədid ‘blow on ’

(from º√puʔ ‘be blown on’ + -igʷəd ‘body’)

dxʷʔəq’yax&adid ‘open  (door)’

(from º√ʔəq’ʷ ‘be open’ + -y-ax&ad ‘side’)

kʷədax&adid ‘take by the arm’

(from √kʷəd ‘be held, be taken’ + -ax&ad ‘arm’)

ləx&šadid ‘light ’s way’

(from √ləx& ‘be light, be bright’ + -šad ‘lower leg’)

ƛ’alšədid ‘put ’s shoes on him’

(from √ƛ’al ‘put  on’ + -šad ‘lower leg’)

ɬič’šadid ‘amputate ’s leg’

(from √ɬič’ ‘get cut with knife’ + -šad ‘lower leg)’

tq’aʔɬdəlid ‘slap  in mouth’

(from √təq’ ‘slap’ + -aʔɬdəɬ ‘mouthpart’

x&əqšadid ‘bind legs of ’

(from º√x&q ‘be wrapped, be tied’ + -šad ‘lower leg’)

Table 46: Stems formed with a lexical suffix and -i-d

With these verbs, however, the effect of -i-d is not applicative but is instead causative, adding an

AGENT/subject to its base rather than adding an object. Consider the examples in (90):

(90) a. ʔucaq’ čəd ʔə tiʔəʔ sx%ədiʔac

ʔu–caq’ čəd ʔə tiʔəʔ sx%ədiʔac

PFV–be.speared 1SG.SUB PR PROX devil’s.club

‘I got speared by the Devil’s Club’

(Bates, Hess & Hilbert 1994: 43)

b. caq’atəbəxʷ ʔə tiʔəʔ caadiɬ tiʔiɬ ʔucutəb ʔə tudiʔ luƛ’ sxʷiʔxʷiʔs əlgʷəʔ

caq’a–t–əb=əxʷ ʔə tiʔəʔ caadiɬ tiʔiɬ ʔu–cut–t–əb ʔə

speared–ICS–PASS=now PR PROX they DIST PFV–speak–ICS–PASS PR

tudiʔ luƛ’ sxʷiʔxʷiʔ–s əlgʷəʔ

DIST.DMA old game–3PO PL

‘what was said by yonder old man to be their game was speared by them’

(Hess 2006: 51, line 224)

c. dxʷcaq’ax%adid tiʔiɬ č’ətx%

dxʷ–caq’–ax%ad–i–d tiʔiɬ č’ətx%

CTD–speared–side–SS–ICS DIST kingfisher

‘he speared Kingfisher in the side’

(Hess & Bates 2004: 20, ex. 71)

(90a) shows the radical, √caq’ ‘be speared, be impaled’, which takes as its syntactic subject the

expression of the PATIENT semantic role and which does not express an AGENT.

In (90b), the

internal causative formed from the same radical, caq’ad ‘spear something’, is shown in its

passive form; here, its subject is the expression of the PATIENT and the AGENT is expressed as an

Cf. the internal causative form of this radical, puʔud ‘blow on something, blow something out’, which nicely

illustrates the contrast in semantic transitivity of the -d and -i-d forms.

The PP ʔə tiʔəʔ sx&ədiʔac ‘by the Devil’s Club’ expresses an inanimate (or at any rate, botanical) INSTRUMENT

rather than an AGENT.

123

oblique object (see Section 6.2 below). The expression of the AGENT in this sentence is allowed

for by the presence of the internal causative suffix -t, which creates a transitive verb from an

intransitive radical (Section 2.1.2.1). Similarly, the form in (90c) is transitive and takes as its

subject the AGENT rather than the PATIENT which is the subject of the radical in (90a). Thus, the

effect on the valency and government pattern of the radical of adding -i-d is the same as that of

adding the transitive causative suffix -t.

This raises the issue of what contribution, if any, the secondary suffix -i- makes to the

meaning of the stem. One possibility is that -i- is not a meaningful element at all and that the

sequence [id] may simply be an allomorph of the internal causative associated with a lexical

suffix (or a particular subset of lexical suffixes). This seems unlikely given that, as we saw in

Section 2.1.6, there are abundant internal causative stems containing lexical suffixes such as

c’agʷačiʔd ‘wash someone’s hands’ (cf. č’əlpačiʔid ‘twist someone’s wrist’ in Table 46) that do

not contain -i-. Another possibility is that the -i- is associated with the possessor-raising seen in

sentences like (90c) whereby it is the bodypart that is affected by the action, but the possessor of

the bodypart is expressed as the direct object. This, however, still begs the question of why -i- is

not present in all transitive stems containing lexical suffixes that express affected bodyparts.

Another reason for not dismissing -i- as a morpheme entirely, at least from a diachronic

perspective, is that it turns up in a few other places as a stem-formative associated with -t. One

particularly suggestive pair of verbs is haʔlid ‘make good for someone, make someone

comfortable’ vs. halʔad ‘tend to someone’, both derived from the adverbial radical √haʔɬ ‘good’.

However, these are the only such contrastive pairs found in the corpus to date, making any

analysis of -i- as anything more than a vestige of an earlier form that was most likely associated

in some way with valency-altering constructions little more than speculation.

124

2.1.4 Incorporative -əɬ

The semantic effects of the suffix -əɬ ‘incorporative [INCRP]’ on the valency of its base are

essentially that of a causative in that it adds a semantic AGENT to a patient-oriented stem;

however, this affix differs from the other causatives discussed above in that its effect on the

syntactic valency of the stem is not to add an object (direct or oblique). Instead, -əɬ increases the

valency of its stem by adding a nominal predicate complement,

as in (91):

(91) a. ʔəstakʷəxʷ

ʔəs–takʷ=əxʷ

STAT–bought=now

‘it’s bought’

(Bates, Hess & Hilbert 1994: 216)

b. ʔutagʷəɬ čəd puʔtəd

ʔu–tagʷ–əɬ čəd puʔtəd

PFV–buy–INCRP 1SG.SUB shirt

‘I bought a/some shirt’

(Hess & Hilbert 1976: II, 138)

Nominal predicate complements like puʔtəd ‘shirt’ in (91) differ from NP arguments of a verb in

that they are not introduced by a determiner and are not referential, but instead have a generic

reference, expressing a type of thing rather than identifying a particular individual or object. The

nominal predicate complement can be made a direct object by adding internal causative suffix -t

to the stem:

(92) siʔ ƛ’ub ɬadsəsƛ’ip’aɬəd čəxʷa ɬuɬčil dxʷdiʔaʔ

siʔ ƛ’ub ɬu=ad=s=ʔəs–ƛ’ip’–aɬ–əd čəxʷa

right.there well IRR=2SG.PO=NM=STAT–compressed–INCRP–ICS 2SG.COORD

ɬu=ɬčil dxʷ–diʔaʔ

IRR=arrive CNTRPT–here

‘right there will you be clutching it and you will arrive here’

(Hilbert & Hess 1977: 19, line 58)

For the syntactic properties of nominal predicate complements, and how these differ from full NP arguments, see

Section 8.2.5)

125

In these construction, -t acts purely as a syntactic transitivizer rather than as a causative.

Although there are few spontaneous examples of -əɬ in the analyzed corpus used in this

grammar, it is nevertheless highly productive in the sense of having wide applicability and can

be combined with any radical that can take the internal causative suffix, -t.

The usual ordering of the nominal predicate complement is immediately post-verbal;

however, it is variably ordered with respect to matrix subject-markers:

(93) a. ʔučaləɬ pišpiš ti sqʷəbayʔ

ʔu–čal–əɬ pišpiš ti sqʷəbayʔ

PFV–chased–INCRP cat SPEC dog

‘the dog chased a/some cat’

(Hess 1995: 120)

b. ʔučaləɬ čəd pišpiš

ʔu–čal–əɬ čəd pišpiš

PFV–chased–INCRP 1SG.SUB cat

‘I chased some/a cat’

(Hess & Hilbert 1976: II, 137)

c. ʔučaləɬ pišpiš čəd

ʔu–čal–əɬ pišpiš čəd

PFV–chased–INCRP cat 1SG.SUB

‘I chased some/a cat’

[based on (93b)]

When in immediate post-verbal position, the integration of the predicate complement to the verb

stem is such that these complements are often considered incorporated elements on a structural

The form ƛ’ip’aɬəd ‘clutch something’ in (92) co-exists with a plain transitive form, ƛ’ip’id ‘clutch something’. It

is not known what semantic difference (if any) there is between the two forms.

There is also one radical in the Lushootseed Dictionary, the adverb √cukʷ ‘very’, that takes -txʷ rather than -t and

also has an -əɬ form, cugʷaɬ ‘be the last ’:

(i) diɬəxʷ ɬucugʷaɬəxʷ

diɬ=əxʷ ɬu=cugʷ–aɬ=əxʷ

FOC=now IRR=very–INCRP=now

‘that will be the last one’

(Bates, Hess & Hilbert 1994: 47)

However, this appears to be a fairly fossilized form, both because it is based on a radical with is (synchronically) an

adverb and because of the idiosyncratic stress pattern [cuˈgʷaɬ], which has prevented the reduction of the vowel in

the incorporative suffix to schwa.

126

par with incorporated nouns, and strings such as ʔučaləɬpišpiš ‘chase cat(s)’ are often written as

single words. Nevertheless, bound enclitics such as =axʷ also readily intervene between the verb

and its complement:

(94) cutəxʷ ti tul’ʔal tiʔiɬ diʔucid daʔəɬəxʷ Vancouver

cut=əxʷ ti tul’ʔal tiʔiɬ diʔ•ucid daʔ–əɬ=əxʷ Vancouver

speak=now SPEC CNTRFG–at DIST other.side•mouth named–INCRP=now Vancouver

‘those from that [place] across the water that is now called Vancouver spoke now’

(Hess 1995: 120)

Nominal predicate complements of stems formed with -aɬ thus do not seem to be fully

incorporated nouns, although they have many of the syntactic and semantic properties of

incorporated nouns in many languages. The degree of integration of the predicate complement in

such structures, however, is particularly interesting from a diachronic point of view, as it

suggests a possible developmental path for lexical suffixes (Section 2.1.6), which may have

evolved from the grammaticization of nouns that were once highly-frequent members of a more

generalized verb + nominal predicate complement type construction.

The incorporative suffix is also found combined with the interrogative word stab ‘what?”

(see Section 8.4.2) to form the interrogative word, stabaɬ ‘what kind?’:

(95) a. stabaɬ titčulbixʷ

stab–aɬ titčulbixʷ

what–INCRP small.animal

‘what kind of little animal is that?’

b. stabaɬ əw’ə qʷɬayʔ tiʔiɬ ʔəsƛ’ax%ʷ ʔal tiʔiɬ

stab–aɬ əw’ə qʷɬayʔ tiʔiɬ ʔəs–ƛ’ax%ʷ ʔal tiʔiɬ

what–INCRP PTCL stick DIST STAT–grow at DIST

‘what kind of wood is that growing over there?

(Bates, Hess & Hilbert 1994: 216)

As with all incorporative stems, stabaɬ takes as a predicate complement a bare noun. This noun

serves to narrow the scope of the question to a particular type of object defined by the nominal

predicate complement. Note that the [-aɬ] allomorph of the suffix in this form is idiosyncratic:

the expected form [stabəɬ] is reported as a variant pronunciation in the Lushootseed Dictionary

127

(Bates, Hess & Hilbert 1994: 216), but [stabəɬ] this is not the phonetic form of the sub-entry

heading nor is it the pronunciation given in the transcribed examples. Hess (p.c.) reports [stabaɬ]

to be the standard pronunciation of this word; whether this reflects the etymological history of

the affix (which also surfaces in this form in čədaɬ ‘which?’ — see Section 2.6.2) will have to

await further comparative work with cognate morphemes in other languages.

2.1.5 Propriative bəs-

The prefix bəs- ‘propriative [PROP]’ is added to nominal bases expressing ‘N’ to create verb

stems meaning ‘have X as one’s N’, as in (96):

(96) a. yəx%i huy ɬubəsčəgʷasəxʷ tiʔəʔ sɬukʷalb ʔə tsiʔəʔ ʔučəbaʔəd tiʔəʔ tatačulbixʷ

yəx%i huy ɬu=bəs–čəgʷas=əxʷ tiʔəʔ sɬukʷalb ʔə tsiʔəʔ

because SCONJ IRR=PROP–wife=now PROX moon PR PROX:FEM

ʔu–čəbaʔ–d tiʔəʔ tatačulbixʷ

PFV–laden–ICS PROX game.animal

‘because Moon will have as his wife she who brought this game’

[DS Star Child, line 305]

b. huy čəxʷ ʔəbsƛ’alabac

huy čəxʷ ʔəs–bəs–s–ƛ’al•abac

SCONJ 2SG.SUB STAT–PROP–NP–put.on•body

‘since you have [them as] clothes’

(Hess 2006: 35, line 325)

c. ʔəbsp’aʔkʷ čəd

ʔəs–bəs–p’aʔkʷ čəd

STAT–PROP–pipe 1SG.SUB

‘I have [it as] a pipe’

(Hess 1998: 30)

The range of nouns which are found with the propriative prefix is restricted to those expressing

kinship, social relations, or things that can be literally owned (as opposed to true possessives,

which can be formed on any noun). Propriative verbs thus express only true ownership and

kinship rather than expressing the vaguer notion of “possession” associated with possessive

affixes.

128

Although the bases are nominal, the derived forms are clearly verbal: they can be marked for

aspect (in most cases, stative aspect) and, when the possessor is first- or second person, subject-

markers are used:

(97) a. gʷəl ɬuʔəbsxʷiʔxʷiʔ čəd ʔə kʷi hiˑkʷ tatačulbixʷ

gʷəl ɬu=ʔəs–bəs–sxʷiʔxʷiʔ čəd ʔə kʷi hikʷ tatačulbixʷ

then IRR=STAT–PROP–game 1SG.SUB PR REM big game.animal

‘and I will have as my game a very large animal’

[MW Star Child, line 74]

b. gʷəl ʔəbsqʷəbqʷəbayʔ əlgʷəʔ ʔə tiʔəʔ bəsaliʔ

gʷəl ʔəs–bəs–sqʷəb–qʷəbayʔ əlgʷəʔ ʔə tiʔəʔ bə=saliʔ

then STAT–PROP–DSTR–dog PL PR PROX ADD=two

‘and they have two (animals) as dogs’

(Hess 1998: 77, line 23)

c. ʔəbsčəgʷas tiʔəʔ bədaʔ ʔə tiʔəʔ sbiaw ʔə tə səsaʔliʔ

ʔəs–bəs–čəgʷas tiʔəʔ bədaʔ ʔə tiʔəʔ sbiaw ʔə tə səsaʔliʔ

STAT–PROP–wife PROX offspring PR PROX coyote PR NSPEC two:HMN

‘Coyote’s son had two wives’ (lit. ‘the son of Coyote has as wives two [people]’)

(Hess 2006: 22)

As shown in (97), propriative stems are bivalent intransitive, the nominal base expressing a

generic type of object (or kin) that defines the relationship and the oblique object further

specifying the possessed. Unlike most oblique objects (Section 8.1.6), however, the object of a

propriative stem is variable in terms of its ordering relative to a subject NP:

(98) a. ʔəbsčəgʷas tiʔiɬ təkʷtəkʷəlus ʔə tsiʔiɬ waq’waq’

ʔəs–bəs–čəgʷas tiʔiɬ təkʷtəkʷəlus ʔə tsiʔiɬ waq’waq’

STAT–PROP–wife DIST owl PR DIST:FEM frog

‘Owl has as his wife Frog’

a. ʔəbsčəgʷas ʔə tsiʔiɬ waq’waq’ tiʔiɬ təkʷtəkʷəlus

ʔəs–bəs–čəgʷas ʔə tsiʔiɬ waq’waq’ tiʔiɬ təkʷtəkʷəlus

STAT–PROP–wife PR DIST:FEM frog DIST owl

‘Owl has as his wife Frog’

(Hess 1998: 31)

In this respect, the possessed resembles an agentive complement (Section 8.1.7); however, unlike

an agentive complement (and unlike oblique objects), the possessed can never be relativized or

used as the predicate in a cleft-construction. Thus, in technical terms, the object of a propriative

129

stem would have to be treated as having a construction-specific grammatical relation to its

syntactic governor.

The propriative prefix bəs- interacts morphophonemically with the stative aspectual prefix

ʔas-, resulting in the fused form [ʔəbs-]. In some cases, the combination ʔas- + bəs- can fuse

with a preceding tense/mood proclitic. Thus, the past tense proclitic tu= combines with ʔas- +

bəs- to give the form [tabs-], while the irrealis proclitic ɬu= combines with these to give [ɬabs-].

However, the alternate forms of both of these sequences of affixes in which the proclitic does not

assimilate — [tuʔəbs] and [ɬuʔəbs], respectively — are also found in texts; in the present corpus

the more highly fused variants are found in the speech of Martha Lamont, a Snohomish speaker,

while the more transparent forms are found in the speech of Skagit narrators (Harry Moses and

Mary Willup). The number of tokens are small, however, and it is possible that the degree of

reduction is dependent on register, style, or rate of speech rather than dialect.

Propriative verb stems can take further derivational affixation:

(99) a. ɬubsčəgʷasbid tsiʔəʔ bədaʔs

ɬu=bəs–čəgʷas–bi–d tsiʔəʔ bədaʔ–s

IRR=PROP–wife–MAP–ICS PROX:FEM offspring–3PO

‘he would marry his daughter’

(Hess 1998: 91, line 6)

b. gʷəl tubəhuyil čəd č’ač’as dəxʷəbsdaʔtubšs ʔə kʷi sp’ip’ic’ikʷ

gʷəl tu=bə=huy–il čəd č’ač’as

then PAST=ADD=be.done–INCH 1SG.SUB child

dəxʷ=ʔəs–bəs–daʔ–txʷ–bš=s ʔə kʷi sp’ip’ic’ikʷ

ADNM=STAT–PROP–named–ECS–1SG.OBJ=3PO PR REM sp’ip’ic’ikʷ

‘and then I was turned into a child so I was named Sp’ip’ic’ikʷ’

[HM Star Child, line 123]

c. x%aƛ’txʷəxʷ kʷi gʷəsəbsč’istxʷils

x%aƛ’–txʷ=əxʷ kʷi gʷə=s=ʔəs–bəs–sč’istxʷ–il=s

desire–ECS=now REM SBJ=NM=STAT–PROP–husband–INCH=3PO

‘she wanted to become someone having a husband’

(Hess 1998: 31)

130

(99a) illustrates the verb bəsčəgʷasbid ‘marry someone’, formed from the stem bəsčəgʷas

‘have someone as a wife’ plus the middle applicative -bi-d. The verb bəsdaʔtxʷ ‘be named

something’ in (99b) is formed by adding the external causative suffix -txʷ to the stem bəsdaʔ

‘have something as a name’, while bəsč’istxʷil ‘become married to someone’ is formed with the

inchoative suffix added to the stem bəsč’istxʷ ‘have someone as a husband’ in (99c).

The

propriative prefix is also found in two kinship terms formed with the prefix ciɬ- ‘half-sibling’ —

ciɬbəsbad ‘half-sibling with same father’ and ciɬbəsk’ʷuy ‘half-sibling with same mother’, based

on the possessive verbs stems bəsbad ‘have a father’ (from √bad ‘father’) and bəsk’ʷuy ‘have a

mother’ (√sk’ʷuy ‘mother’), respectively.

2.1.6 Other verb-stem formatives

-a- ‘stem formative’

The stem-formative element -a- is identified by Hess (1967a: 34) and Hess & Hilbert (1976:

II, 152) as appearing in a number of verb stems with rather uncertain effects on the meanings of

its base. It appears in a few verb forms in tandem with the valency-neutral middle suffix -b —

e.g., ckʷab ‘be taut’ (from √cikʷ ‘be straight, be tautened’) and qʷcab ‘slip’ (from √qʷc ‘slide,

slip’)

— and Hess (1967a) suggests that it may also be a component of the suffix -áb ‘method’

(Section 2.1.1.6), used to derive verbs from nouns. Two motion verbs, ʔəƛ’ ‘come’ and ʔux&ʷ

‘go’, also have forms in /a/ — ʔəƛ’ ‘come to a specific place’ and ʔux&ʷ ‘go to a specific place’,

and both Hess (1967a: 34) and Hess & Hilbert (1976: II, 152) suggest that the stem-formative /a/

is a conditioning factor for the syncope of the instrumental suffix -təd in words like ƛ’əgʷiʔčad

The impersonal-object reading of the example itself is allowed for by the fact that the possessum is an optional

argument of the verb which may be left unspecified to be recovered from context or interpreted as a generic

thing/person (depending on the meaning of the verb stem).

These might also include ɬx&ab ‘be stiff (as a result of an encounter with a spirit)’, possible from º√ɬx& ‘be spread

out’, although the derivational relationship between these forms is uncertain.

131

‘spear for bottom-fishing’ (from ƛ’agʷiʔč ‘spear something (bottom fish)’) and yiq’ibad ‘awl for

basket-making’ (from yiq’ib ‘make something (baskets)’). Whether or not all of these /a/ vowels

constitute a morpheme or morphemes, however, is quite uncertain — from a synchronic

perspective all of these examples of /a/ seem to be unanalyzable parts of fossilized stems. The

extent to which these stems can be teased apart to reveal a historical *-a- affix (or affixes) seems

like a promising topic for future investigation.

gʷə- ‘dubitative’

There are a few verb stems that contain the prefix gʷə- ‘dubitative [DUB]’, which adds an

element of doubt or uncertainty to the meaning of its base — e.g., gʷəhaydxʷ ‘sort of know

something’ (from haydxʷ ‘know something’). In these cases, the prefix is most likely a

lexicalization of the subjunctive proclitic (Section 3.1.4) as a part of the verb stem. Less clearly

related to these stems are verbs such as gʷəbak’ʷtxʷ ‘catch a glimpse of something’ (probably

from √bak’ʷ ‘move rapidly’), gʷəbiƛ’ad ‘make something disintegrate’ (from º√biƛ’ ‘be

smashed, be crumbled’), gʷəcutad ‘make noise’ (based on √cut ‘speak’), gʷədᶻakʷtxʷ ‘be

emotionally moved’ (probably from dᶻakʷ ‘be shaky’), gʷəƛ’əlad ‘stop, be still; behave’ and

gʷəƛ’əltxʷ ‘stop someone’, gʷəqʷibad ‘get ready’ (based on √qʷib ‘be fixed’), gʷəšəbad ‘make

something disappear’ (based on √šub ‘be overdue, be missing’), and gʷət’q’ʷad ‘faint, pass out’

(probably from √t’q’ʷ ‘be broken (flexible object)’). These verbs are only partially analyzable

and the semantic contribution of the dubitative prefix (if indeed it appears in these forms) is far

from apparent. Note that the several of these forms also appear to contain the stem-formative,

-a-.

2.2 Nouns and nominal derivation

In comparison with verbs, nouns in Lushootseed do not have a particularly complicated

morphological structure, nor are there a great many fully productive derivational affixes used to

132

derive nouns from words of other lexical classes. However, as noted in Hess (1967a), many

nominal bases are phonologically more complex than the majority of verbal radicals and show

signs of once having been composed of morphological elements no longer at play in the modern

language. Indeed, a close look at the nominal lexicon gives the impression that at one stage of

the language’s history most nouns had complex morphological structure, and even in the modern

language a substantial portion of nouns can be analyzed as deverbal, having been formed with

the nominalizing prefix s- (Section 2.2.1). Similarly, an examination of the syntax reveals a

preponderance of syntactic strategies for creating referential expressions (the syntactic

equivalents of NPs) from verbal and other predicative phrases (see Section 7), hinting at an

earlier stage of the language with a much more verbally-oriented lexicon with few or no basic

nouns, perhaps more akin to what is still seen in the Iroquoian family today. The sections that

follow describe the synchronically-analyzable noun-deriving affixes in Lushootseed, beginning

with the most productive morphemes and ending with a brief enumeration of some fossilized

elements which are nevertheless still analyzable from an etymological point of view.

2.2.1 Nominalizing prefix s-

The prefix s- ‘nominalizing prefix [NP]’ (not to be confused with the nominalizing proclitic,

s=, see below) is the most common and prolific of the derivational affixes in Lushootseed. Its

primary function is to create lexical nouns from verbal bases, as in the forms in Table 47:

sc’əp’ ‘swamp’

(√c’əp’ ‘be stagnant’)

sdukʷ ‘worthless person; supernatural thing’

(√dukʷ ‘be worthless; be anormal’)

sgʷaadgʷad ‘tape recorder’

(√gʷaadgʷad ‘converse’)

səliʔ ‘soul’

(√həliʔ ‘be alive’)

sləx&il ‘day’

(º√ləx& ‘be light, be bright’; cf. ləx&il ‘become light’)

sƛ’iqabac ‘small pox’

(from √ƛ’iq ‘emerge’ + -abac ‘body’)

spukʷəb ‘hill’

(√pukʷəb ‘be piled up’)

sp’qʷucut ‘spawning steelhead’

(from √p’q’ʷ ‘drift’)

sqəlalitut ‘spirit power’

(from qəlalitut ‘dream’)

sq’əčqs ‘coho salmon’

(from √q’əč ‘be crooked’ + -qs ‘nose’)

sq’čic ‘bow (archery)’

(from √q’əč ‘be crooked’ + -ič ‘spine’)

sqʷicqs ‘point (land)’

(from √qʷic ‘go downstream’ + -qs ‘nose’)

sqʷšab ‘fog’

(√qʷšab ‘be foggy’)

133

sq’ʷic’ ‘widow, widower’

(º√q’ʷic’ ‘be a widow(er)’; cf. q’ʷic’il ‘be widowed’)

stibtib ‘strong person’

(√tib ‘make great physical effort’)

studəq ‘slave’

(º√tudəq ‘be a slave’; cf. tudəqil ‘be enslaved’)

st’q’ʷalšəd ‘tumpline’

(from √t’q’ʷ ‘be snapped in two’ + -šəd ‘lower leg’)

sx&aʔx&aʔ ‘in-law’

(√x&aʔx&aʔ ‘be taboo’)

sx&əɬ ‘sickness’

(√x&əɬ ‘be sick’)

sx&ʷəs ‘fat, grease’

(º√x&ʷəs ‘be fat’; cf. x&ʷsil ‘get fat’)

Table 47: Nouns formed with s- from monovalent bases

The forms in this table fall roughly into two groups. The first are nouns such as sdukʷ ‘worthless

person; supernatural thing’ (from √dukʷ ‘be worthless; be a-normal’) or studəq ‘slave’ (º√tudəq

‘be a slave’) that express the ACTOR/subject of the verb it is derived from. The second group

consists of nouns that express a resultant state of the process expressed by the verb — e.g., sx&əɬ

‘sickness’ (from √x&əɬ ‘be sick’) or sqəlalitut ‘spirit power’ (from qəlalitut ‘dream’). In both

cases the meanings of many of the nouns are lexicalized such that they express a particular

ACTOR/subject or resultant state out of the potential range of things that a completely productive

nominalization might express — i.e., sq’əčqs (from √q’əč ‘be crooked’ + -qs ‘nose’) means

‘Coho salmon’, not anything with a crooked nose, and sgʷaadgʷad (from √gʷaadgʷad

‘converse’) means ‘tape recorder’, not any thing or person that can speak. There are also cases

where the relationship between the meaning of the noun and the meaning of its base seems to be

a matter of loose semantic association than regular semantic derivation (for example, səliʔ ‘soul’

from √həliʔ ‘be alive’, sx&ʷəs ‘fat, grease’ from º√x&ʷəs ‘be fat’). The relationship between the

meaning of the noun and the meaning of its verbal bases may also be rather oblique (e.g.,

st’q’ʷalšəd ‘tumpline’ from √t’q’ʷ ‘be snapped in two’ + -šəd ‘lower leg’, a reference to the way

tumplines were traditionally made) or completely opaque (sqʷicqs ‘point (land)’ from √qʷic ‘go

downstream’ + -qs ‘nose’).

There are also a great many nouns formed from bivalent intransitive bases. In these cases, the

derived noun expresses the semantic ENDPOINT/oblique object of its base:

sʔuladxʷ ‘anadromous fish’

(º√ʔuladxʷ ‘catch  [anadromous fish]’)

sʔax&ʷuʔ ‘clam’

(√ʔax&ʷuʔ ‘catch  [clams]’)

sʔəɬəd ‘food’

(√ʔəɬəd ‘feed on ’)

134

sʔuləxʷ ‘things collected from nature’

(√ʔuləxʷ ‘collect from nature’)

sčəbaʔ ‘backpack’

(√čəbaʔ ‘be loaded down with ’)

sdaʔ ‘name’

(º√daʔ ‘be named ’; daʔad ‘name ’)

sgʷadadgʷad ‘story’

(√gʷadadgʷad ‘speak ’)

sɬub ‘soup’

(º√ɬub ‘eat  with spoon’; cf. ɬubtxʷ ‘spoon feed ’)

st’ilib ‘song’

(√t’ilib ‘sing ’)

sxʷiʔxʷiʔ ‘food gathered by hunting or foraging’

(√xʷiʔxʷiʔ ‘hunt for , forage for ’)

Table 48: Nouns formed with s- from bivalent intransitive bases

In most of these examples, the meaning of the derived noun is predictable and completely

transparent; however, there are a few examples — like sɬub ‘soup’ (from º√ɬub ‘eat something

with a spoon’) — where the noun seems to have lexicalized to a more specific meaning. Note

that there are no examples of this type of nominal derivation based on transitive verbs: transitive

verbs are nominalized on the clausal-level and require expression of a subject (see Section 7.4.2

below).

Nouns formed with s- can themselves be used as bases for deriving other words. In these

cases, the s-prefix is usually lost, as in the following examples:

(100) a. ʔal kʷi tuhaʔkʷ gʷəl xʷiʔ gʷətusləx%il

ʔal kʷi tu=haʔkʷ gʷəl xʷiʔ gʷə=tu=(s–ləx%–il)

PR REM PAST=long.time then NEG SBJ=PAST=(NP–light–INCH)

‘in the distant past there was no daylight’

(Hilbert & Hess 1977: 13, line 1)

b. ləx%iličəxʷ čəɬ

˹ləx%–il˺•ič=əxʷ čəɬ

˹light–INCH˺•covering=now 1PL.SUB

‘daylight is covering us’

(Hilbert & Hess 1977: 30, line 146–147)

The first example shows the noun sləx&il ‘day, daylight’, derived from the inchoative form of the

verbal radical √ləx&il ‘be light, be bright’. In (100b), this noun appears as the base for a verb

derived through the affixation of the lexical suffix, -ič ‘covering’; the fact that the noun, rather

than the verbal radical, is the base for this form is shown by the relative ordering of the affixes

— lexical suffixes are attached directly to the stem, followed by other derivational affixes,

The valency and government patterns of these forms are uncertain.

135

including the inchoative (cf., the form cəbagʷiɬil [cəb ‘two’ + -agʷiɬ ‘canoe’ + -il ‘inchoative’]

‘become two canoes’).

A few nouns formed with s- seem to be based on nominal stems:

sʔacus ‘face’

(from º√ʔac ‘centre’ + -us ‘face’)

sč’ətšad ‘fish tail’

(from º√č’ət ‘point’ + -šad ‘lower leg’)

sɬagʷidac ‘bed sheet’

(√ɬagʷid ‘sleeping mat’)

sčəbid ‘Douglas fir bark’

(√čəbid ‘Douglas fir’)

sʔilax&ad ‘side of something’

(from ʔilax&ad ‘side’ based on √ʔil ‘be leaning on’ + -ax&ad ‘side’)

sʔilgʷiɬ ‘shoreline’

(from ʔilgʷiɬ ‘shoreline’ based on √ʔil ‘be leaning on’ + -gʷiɬ ‘canoe’)

sdᶻabid ‘vegetables’

(√dᶻabid ‘vegetables’)

st’əluʔb ‘dried king salmon’

(t’əluʔb ‘dried king salmon’ from º√t’al ‘be split open’)

Table 49: Nouns formed with s- from nominal bases

Several of these forms, such as sdᶻabid ‘vegetables’ (from √dᶻabid ‘vegetables’) and st’əluʔb

‘dried king salmon’ (from t’əluʔb ‘dried king salmon’), are homosemous with their bases,

possibly representing forms in which the prefix is being eroded by diachronic processes. In other

cases, the nominalization seems to represent a narrowing of scope — sɬagʷidac ‘bed sheet’

(from √ɬagʷid ‘sleeping mat’), sčəbid ‘Douglas fir bark’ (from √čəbid ‘Douglas fir’) — from

the meaning of the base. Two of the forms in Table 49 — sʔacus ‘face’ and sč’ətšad ‘fish tail’

— are formed by the combination of a lexical suffix (Section 2.1.6) with a bound radical

expressing a partonymic notion, and two more — sʔilax&ad ‘side of something’ and sʔilgʷiɬ

‘shoreline’ — are based on the verbal radical √ʔil ‘be leaning on’ plus a lexical suffix. A number

of less transparent nouns, such as sc’ic’alb ‘wing’ (cf. √c’ic’al ‘quill feather’), also seem to have

nominal bases (or historically to share bases with other nouns).

Finally, there are very many nouns in Lushootseed that begin with /s/ but are synchronically

unanalyzable. A few of these are given in Table 50:

sʔəx&ʷaʔ ‘male urine’

sq’čic ‘bow (archery)’

ʔic’əb ‘blanket’

sq’ədᶻax& ‘intestines’

sbadil ‘mountain’

sq’ədᶻu ‘human hair’

sbəq’ʷaʔ ‘heron’’

sqa ‘older sibling or cousin’

sbiaw ‘coyote’

sqigʷac ‘deer’

sbit’ ‘soup’

sqʷaliʔ ‘hay’

sc’aliʔ ‘heart’

sqʷəbayʔ ‘dog’

sč’ətx& ‘kingfisher’

sqʷiƛ’əb ‘spear for bottom fishing’

sč’istx& ‘husband’

st’ət’tqʷiʔ ‘drake bufflehead’

136

scapaʔ ‘grandfather’

st’uʔq’ʷ ‘small feathers’

sčətxʷəd ‘black bear’

stigʷəd ‘cedar withes’

sčutx& ‘halibut’

stiqayuʔ ‘wolf’

sdəxʷiɬ ‘hunting canoe’

stiqiw ‘horse’

sgʷaʔɬtəd ‘funeral goods’

stubš ‘man’

sgʷəlub ‘pheasant’

stuləkʷ ‘river’

skʷup ‘sucker fish’

stuligʷəd ‘blood’

skʷxʷic ‘silver salmon’

swatixʷtəd ‘land, region, world’

sɬadəyʔ ‘female human’

swuqʷadiʔ ‘loon’

sɬukʷalb ‘moon’

sx&aʔhus ‘big river sawbill’

sɬuʔb ‘dog salmon’

sx&ədᶻəb ‘dried camas root’

spaɬx&ad ‘tidal flats’

sx&əy’us ‘head’

spaʔc ‘black bear’

sx&ʷədiʔ ‘bullhead’

spčuʔ ‘watertight basket’

syaʔyaʔ ‘friend’

sq’aʔšəd ‘moccasin’

syəhub ‘traditional story’

Table 50: Unanalyzable nouns with s-

Although none of the initial /s/’s in these words can be analyzed as a synchronically

productive application of the nominalizing prefix, their diachronic origins are revealed by the

absence of /s/ when the noun is used as a base for the derivation of other forms — as, for

example, in skʷxʷic ‘silver salmon’ > pədkʷxʷic ‘September’ (lit. ‘time of silver salmon’),

sbəq’ʷaʔ ‘heron’ > bəq’ʷaʔqs ‘bayonet’ (lit. ‘heron nose’), or sbiaw ‘coyote’ > biawcut ‘act

cannily’. The words sɬadəyʔ ‘woman’ and stubš ‘man’ lose /s/ in the formations ɬadəyʔlucidbid

‘address as a female’ and tubšlucidbid ‘address as a male’, and the kinship terms sk’ʷuy

‘mother’ and scapaʔ ‘grandfather’ both lose their initial /s/ to form vocatives. This may also be

related to the phenomenon found in Lushootseed texts whereby proper names are often prefixed

with s- when not used as vocatives This gives us expressions such as (101):

The apparent radical of this form is attested in at least one example in the texts:

(i) ʔəsɬukʷalb tiʔiɬ suqʷaʔs

ʔəs–ɬukʷalb tiʔiɬ suqʷaʔ–s

STAT–be.moon DIST younger.brother–3PO

‘his younger brother was the moon’

[DS Star Child, line 235]

Although this verbal radical is not attested elsewhere, a further hint of a verbal origin for the word sɬukʷalb ‘moon’

is its similarity to ɬukʷaɬ ‘sun’. Further etymological and comparative work will likely shed some light on the

origins and historical relationships of these words.

137

(101) ləkʷatač dxʷʔa, tiʔiɬ sƛ’iƛ’iq’šəd

lə=kʷatač dxʷ–ʔa tiʔiɬ s–ƛ’iƛ’iq’šəd

PROG=climb CNTRPT–be.there DIST NP–sapsucker

‘Sapsucker was climbing there’

(Hess 2006: 24, line 47)

However, it is also the case that names formed with s- occasionally appear without the prefix in

non-vocative usages as well, so if there is a relationship between the two processes, it is a

historical one that has given rise to a certain amount of free variation in the modern language.

The initial /s/ in all nouns like those in Table 50 is ignored in reduplication (Section 2.8), which

may be a vestige of a time when these words were analyzable s-nominalizations.

The final issue that merits discussion here is the relationship between the nominalizing prefix

s- and the nominalizing proclitic s= (see also Section 7.4.2). While these are almost certainly

cognate elements historically and have traditionally been treated as the same element in

Lushootseed as well as in other Salishan languages, careful consideration of their behaviour and

distribution shows clearly that they are formally distinct morphemes. The nominalizing prefix

operates at the level of the word, creating new lexemes and, in most cases, effecting a change in

lexical class; the nominalizing clitic operates at the level of the phrase, creating referential

expressions from full clauses that are the syntactic equivalent of NPs. The s- prefix attaches

directly to the radical and is closer to its base that other derivational prefixes; the proclitic s= is

bound not to a word of a particular class but appears attached to the first element in the phrase it

is associated with and, like other proclitics, may be iterated on a subsequent element in that

phrase. The nominalizing prefix is restricted to intransitive bases, while the proclitic can be

applied to a predicative phrase headed by a predicate of any valency. As a word-level

derivational morpheme, s- has no effect on the syntactics of its base other than its basic lexical

class-changing function of creating a noun; the s= proclitic, on the other hand, applies to

predicate phrases headed by a fully inflected sentence predicate (usually a verb) and requires that

its phrase contain some expression of its syntactic subject in the form of a possessive subject

138

clitic or a possessor NP. Thus, while the parallels between s- and s= are undeniable, their

syntactic behaviour is too disparate to allow them to be treated as precisely the same element.

2.2.2 Proclivitive dxʷs-

The prefix dxʷs- ‘proclivitive [PROC]’ is added to verbal bases to form words denoting one

whose profession or proclivity is to perform a particular action:

dxʷsʔəɬəd ‘one who eats a lot’

(√ʔəɬəd ‘eat ’)

dxʷscutcut ‘one who talks a lot’

(√cut ‘speak’)

dxʷčəbaʔ ‘porter’

(√čəbaʔ ‘be loaded down with ’)

dxʷsbədč ‘liar’

(º√bədč ‘lie’; cf. bədčəb ‘tell lies’)

dxʷskʷatač ‘mountain climber’

(√kʷatač ‘climb’)

dxʷsp’ayəq ‘canoe maker’

(√p’ayəq ‘carve canoe’)

dxʷsqadaʔ ‘thief’

(√qadaʔ ‘steal ’)

dxʷsqʷɬalps ‘drover’

(from √qʷaɬ ‘drive  [animal]’ + -alps ‘animal’)

Table 51: Nouns formed with dxʷs-

To this list could also be added the synchronically unanalyzable form dxʷsʔubədiʔ ‘forest

hunter’ which is based on a root which is unattested except in the synonymous form sʔubədiʔ

‘forest hunter’.

Although they are ordinary nouns in other respects, nouns formed with dxʷs- do not normally

take possessive affixes. Instead, when a possessive might be required, proclivitive nouns

combine with the instrumental prefix səxʷ- (Section 2.2.3) to form special possessive

constructions like that in (102):

(102) diɬ acəxʷdxʷsp’ayəq

diɬ ad–səxʷ–dxʷs–p’ayəq

FOC 2SG.PO–INST–PROC–carve.canoe

‘he is your canoe-maker’

(Hess 1971: 52)

Although it may seem undesirable to posit two homophonous and (superficially) homosemous elements as

separate morphemes, it is by no means without parallel in other languages, even in familiar languages from the Indo-

European family. Igor Mel’čuk (personal communication) notes that this is the case in Russian, where many

prepositions have homophonous prefixal counterparts that add meanings to verbal bases that are not unrelated to the

meanings of the independent prepositions. One might make the same argument for some prepositions in English,

which have homophonous counterparts in phrasal verbs that are syntactically very unlike prepositions, but which

often express meanings that are semantically-related to the meanings of the corresponding preposition.

139

The səxʷ- prefix otherwise attaches only to verbal bases.

As noted in Hess (1995: 25 – 28), the proclivitive is almost certainly related diachronically to

the prefix dxʷ(s)- ‘contained’ (Section 2.1.1.1), which expresses the general notion of internality

or containment. It probably also contained the nominalizing prefix s- (Section 2.2.1) added to a

verbal radical to create a nominal of activity, the potential or origin of the forms in Table 51

being expressions with a literal meaning of ‘one whose has internalized an activity’. Also like

dxʷ(s)- ‘contained’, dxʷs- ‘proclivitive’ is occasionally attested as [dxʷ-] or [xʷ-], although there

are no such attestations in the present corpus.

2.2.3 Instrumental səxʷ-

The prefix səxʷ- ‘instrumental [INSTR]’ is added to verbal bases to create nouns expressing an

instrument that is habitually used to perform the action or activity expressed by its stem. This

affix interacts morphophonemically with the first- and second-person singular possessive

prefixes:

(103) a. cəxʷp’ac’ b. acəxʷp’ac’

d–səxʷ–p’ac’ ad–səxʷ–p’ac’

1SG.PO–INSTR–sew 2SG.PO–INSTR–sew

‘my sewing machine’ ‘your sewing machine’

(Hess 1971: 50)

This reduction results in homophony with forms bearing the adjunct nominalizing proclitic

dəxʷ= (Section 7.4.2.2) with first-person singular possessive subjects (compare 103a and

cəxʷp’ac’ ‘why I sew’).

The prefix səxʷ- appears to be quite productive and appears in a large number of forms. A

selection of these are given in Table 52:

səxʷʔax&ʷuʔb ‘clam gun’

(ʔax&ʷuʔb ‘dig for clams’)

səxʷʔigʷəɬ ‘ladder, stairway’

(√ʔigʷəɬ ‘climb’) (Sk)

səxʷʔix&ʷicut ‘broom’

(ʔix&ʷicut ‘throw oneself’)

səxʷʔulʔul ‘means for water travel’

(√ʔulʔul ‘travel by water’)

səxʷc’əgʷulč ‘dishwasher’

(from √c’akʷ ‘wash ’ + -ulč ‘container’)

səxʷc’ix&alikʷ ‘frying pan’

(c’ix&alikʷ ‘fry ’)

140

səxʷgʷədil ‘chair’

(from gʷədil ‘sit down’ based on √gʷəd ‘down’)

səxʷkʷatač ‘ladder; upwards trail’

(√kʷatač ‘climb’)

səxʷɬaq ‘spinning wheel’

(√ɬaq ‘spin wool’) (Sk)

səxʷɬič’ ‘saw’

(√ɬič’ ‘be cut’)

səxʷƛ’akʷ ‘thing used to make mats’

(º√ƛ’akʷ ‘be stitched’; cf. ƛ’agʷəd ‘stitch  (mat)’)

səxʷp’ac’ ‘sewing machine’

(√p’ac’ ‘sew ’)

səxʷšuɬ ‘spyglass’

(√šuɬ ‘appear, be visible’)

səxʷsx&aysəb ‘electric razor’

(from √sx& ‘scrape’ + -ayqs ‘chin’ + -b ‘middle’)

səxʷuc’əgʷulč ‘thing used to wash dishes’

(from √c’akʷ ‘wash ’ + -ulč ‘container’)

səxʷuɬaq ‘thing used to spin wool’

(√ɬaq ‘spin wool’) (Sk)

səxʷup’ac’ ‘thing used to sew’

(√p’ac’ ‘sew ’)

səxʷuyayus ‘tool’

(√yayus ‘work’)

səxʷx&al ‘writing implement’

(√x&al ‘write’)

səxʷx&ix&q’alps ‘race horse’

(from √x&ix&q’ ‘be competitive’ + -alps ‘animal’)

Table 52: Nouns formed with səxʷ-

To this list might also be added səxʷsʔuləx& ‘foraging territory’ (from √ʔuləx& ‘forage for

something’), although the presence of the nominalizing prefix s- and the form’s unusual

semantics (referring to a place rather than an instrument) argues for a more complex

etymological history. There is also a kinship term, səxʷsqatəd ‘older siblings, older cousins’

(from √sqa ‘older sibling, older cousin’), which appears to contain both səxʷ- and the implement

suffix -təd, although the meaning of neither of these affixes is apparent in the meaning of the

form.

While many of the forms in Table 52 have become lexicalized to express specific items used

in certain activities (e.g., səxʷgʷədil ‘chair’, səxʷp’ac’ ‘sewing machine’), several of these forms

have a more general sense and are applicable to any instrument used in an activity (səxʷup’ac’

‘thing used to sew’, səxʷƛ’akʷ ‘thing used to make mats’). This is in contrast to the implement

suffix -təd (Section 2.2.9), which creates nouns referring to “specific, well-defined items of

traditional shape and established usage” (Hess 1971: 47). The distinction between the two affixes

can be seen in a handful of forms that take either or both of səxʷ- and -təd:

(104) səxʷp’ac’təd p’ac’təd səxʷp’ac’

‘thing used to sew’ ‘sewing needle’ ‘sewing machine’

səxʷɬaqtəd ɬaqtəd səxʷɬaq

‘thing used to spin yarn’ ‘leg spindle’ ‘spinning wheel’

141

səxʷƛ’akʷtəd ƛ’akʷtəd səxʷƛ’akʷ

‘mat-making needle’ ‘cattail needle’

‘things used for making mats’

səxʷčəbaʔtəd čəbaʔtəd —

‘thing used for carrying on back’ ‘tumpline’ —

As shown in the first triplet in (104), adding the instrumental prefix to a noun formed with -təd

has the effect of making the expression less specific: rather than expressing a particular item

(p’ac’təd ‘sewing needle’) used in an activity (√p’ac’ ‘sew’), the derived noun (səxʷp’ac’təd

‘thing used to sew’) designates any item used in that activity. The remainder of the examples

illustrate the same point: in each case the noun bearing the instrument prefix and -təd is less

specific than the noun bearing only -təd. In three of the four cases, there are also contrasting

forms that contain only səxʷ-. While one of these, səxʷƛ’akʷ ‘things used for making mats’, has

the usual open-ended meaning associated with səxʷ-, two of these — səxʷp’ac’ ‘sewing

machine’ and səxʷɬaq ‘spinning wheel’ (as well as a number of the other forms in Table 52) —

designate modern devices or machines used to perform an activity rather than the traditional

implement designated by the nouns formed with -təd. The fact that so many of words of this type

are of recent origin suggests that they are descriptive expressions that became lexicalized as the

names of new or novel items originally identified by their function. It should also be noted that

both səxʷp’ac’ ‘sewing machine’ and səxʷɬaq ‘spinning wheel’ contrast with more open-ended

forms (səxʷup’ac’ ‘thing used to sew’ and səxʷuɬaq ‘thing used to spin wool’, respectively)

which appear to be based on verb stems inflection for the perfective aspect (ʔu-). In addition,

there is a form səxʷuƛ’akʷtəd ‘thing used to make mats’ (cf. səxʷƛ’akʷtəd ‘mat-making needle’)

which also appears to contain the perfective marker, and another form, səxʷəsq’ʷuʔil ‘device for

keeping what has been collected, book-binder’, which seems to be based on a stem bearing the

A cattail needle was a specific type of implement, about four feet long and made from ironwood, used only for

making cattail mats.

142

stative aspect (ʔəs-).

If these are indeed based on stems inflected for aspect, this seems to

suggest that səxʷ- may have been applicable not just to verb stems but also to full clauses, much

like the morphemes s= and dəxʷ= discussed in Section 7.4.2 below. However, it is difficult to

speculate on the basis of fewer than half a dozen forms, and the question will have to be left open

for future research.

səxʷ- is also found on stems formed with the proclivitive, dxʷs- (Section 2.2.1), as in (105):

(105) diɬ acəxʷdxʷsp’ayəq

diɬ ad–səxʷ–dxʷs–p’ayəq

FOC 2SG.PO–INST–PROC–carve.canoe

‘he is your canoe-maker’

(Hess 1971: 52)

As noted above, nouns formed with the proclivitive do not normally take possessive affixes. This

use of the instrumental is also unusual in that otherwise this prefix is restricted to verbal bases.

2.2.4 Proxy sixʷ-

The prefix sixʷ- ‘proxy’ is attached to verb stems to form nouns designating a person who

serves in a particular capacity on behalf of or at the behest of someone else. Like səxʷ-, it

undergoes morphophonemic reduction when preceded by the first- and second-person singular

possessive prefixes:

(106) a. cixʷup’ayəq b. acixʷup’ayəq

d–sixʷ–ʔu–p’ayəq ad–sixʷ–ʔu–p’ayəq

1SG.PO–PROXY–PFV–carve.canoe 2SG.PO–PROXY–PFV–carve.canoe

‘my canoe-maker’ ‘your canoe-maker’

(Hess 1971: 50)

This is a low-frequency prefix and is attested in only five forms: cixʷup’ayəq ‘one’s canoe-

maker’ (from √p’ayəq ‘carve canoe’), cixʷstudəq ‘one’s slave’ (from √studəq ‘slave’),

cixʷc’əgʷulč ‘one’s dish-washer’ (from √c’akʷ ‘wash something’ + -ulč ‘container’),

The stem itself, q’ʷuʔil, is unattested, although it is probably based on q’ʷuʔ ‘be gathered’ plus the inchoative

suffix -il.

143

cixʷudxʷc’əgʷulč ‘one’s dish-washer’ (from dxʷs- ‘proclivitive’ + √c’akʷ ‘wash something’ +

-ulč ‘container’), and sixʷsyayaʔ ‘in-laws’ (from s√yayaʔ ‘relative, friend’). Two of these —

cixʷup’ayəq and cixʷudxʷc’əgʷulč — appear to contain the perfective prefix ʔu-, and so may be

based on the nominalization of clauses rather than verb stems; however, with only four attested

forms, any conclusions that might be drawn from this are little more than speculation.

2.2.5 Container xʷ-

The prefix xʷ- ‘container [CNTNR]’ is added to both nominal and verbal roots, usually in

conjunction with some other affix, to create nouns expressing various types of containers. This

affix is notable for its morphophonemic interaction with the first- and second-person possessive

markers, as shown in (107):

(107) a. cxʷtaləhali b. acxʷtaləhali

d–xʷ–talə•hali ad–xʷ–talə•hali

1SG.PO–CNTNR–money•place 2SG.PO–CNTNR–money•place

‘my purse’ ‘your purse’

(Hess 1971: 53)

As noted by Hess (1971), this interaction probably indicates that historically xʷ- has its origins in

the sequence *s-xʷ-, the /s/ being lost by assimilation to the following fricative. The prefix xʷ- is

also occasionally dropped by speakers, as in the following example:

(108) dəgʷdəgʷašəxʷ ʔal tiʔəʔ ʔuhuyud əlgʷəʔ sdəgʷigʷsalis

dəgʷ–dəgʷa–š=əxʷ ʔal tiʔəʔ ʔu–huyu–d əlgʷəʔ

DSTR–inside–ICS=now at PROX PFV–be.done–ICS PL

s=dəgʷ•igʷs•ali=s

NM=inside•things•place=3PO

‘they put them in their pockets as they made them’

[DS Star Child, line 337]

The expected from for ‘pocket’ in (108) is xʷdəgʷigʷsali; however, form on tape is

dəgʷigʷsaliʔ. As noted in Hess (1971; 1995), such elision is not infrequent, and so far efforts to

unravel its conditioning have failed.

144

A number of transparent examples that illustrates this prefix’s most regular meaning,

‘container’, are given in Table 53:

xʷc’agʷačid ‘wash basin’

(√c’akʷ ‘wash ’ + -ačiʔ ‘hand’ + -təd ‘implement’)

xʷciqəbad ‘ramrod’

(from xʷ- ‘container’ + √ciq ‘be impaled’ + -b ‘middle’)

xʷdᶻədisali ‘gums’

(√dᶻədis ‘tooth’ + -ali ‘place’)

xʷdəgʷigʷsali ‘pocket’

(√dəkʷ ‘be inside’ + -igʷs ‘things’ + -ali ‘place’)

xʷgʷədgʷatəd ‘voice; language’

(√gʷad ‘talk’ + -təd ‘implement’)

xʷƛ’aləp ‘pot’

(√ƛ’al ‘fit within’ + -ap ‘bottom’)

xʷlabali ‘bottle’

(√lab ‘rum’ + -ali ‘place’)

xʷlikəlihali ‘lock’

(√likəli ‘key’ + -ali ‘place’)

xʷqʷuʔəd ‘bucket’

(√qʷuʔ ‘water’ + -təd ‘implement’)

xʷsx&ʷaʔəd ‘bladder’

(√sx&ʷaʔ ‘male urine’ + -təd ‘implement’)

xʷt’agʷtəp ‘chair’

(√t’agʷt ‘be on top’ + -ap ‘bottom’)

xʷtaləhali ‘purse’

(√talə ‘money’ + -ali ‘place’)

Table 53: Nouns formed with xʷ-

All of the examples in Table 53 contain either the instrument suffix -təd (Section 2.2.6) or the

lexical suffix -ali. To this list we could add a number of less transparent forms containing xʷ-

such as xʷʔax&ʷaʔəd ‘clam basket’ (possibly from √sʔax&ʷuʔ ‘butter clam’ + -təd ‘implement’),

xʷqəʔltəd ‘diaper’, xʷšiʔšɬəbəd ‘window, mirror’ (based on √šuɬ ‘appear, be visible’), xʷč’up’c’

‘tail’, xʷč’iɬqs ‘deepwater oyster shell’, and xʷwiyus ‘mask’, as well as a few lexicalized

expressions like xʷcqʷuɬ ‘day; sun at zenith’ (from √cqʷuɬ ‘post’), xʷqʷq’ʷus ‘cliff’ (from

√qʷq’ʷ ‘white’ + -us ‘face’), xʷɬəbc’qs ‘mucus’ (from º√ɬəbc’ ‘mucus’ + -qs ‘nose’), xʷsqatəd

‘older siblings, older cousins’ (from √sqa ‘older sibling, parent’s older sibling’ + -təd

‘implement’), and xʷšəbus ‘partly dried fish’ (from √šab ‘be dry’ + -us ‘face’).

The prefix xʷ- is also found in a number of bodypart terms, including xʷtabəp ‘rump’

(contains -ap ‘bottom’), xʷčɬičəp ‘sacrum’ (contains -ič ‘covering’ and -ap ‘bottom’), xʷʔiləx&əd

‘side of body’ (from √ʔil ‘edge’ + -ax&ad ‘side’), xʷʔəcušəd ‘sole of foot’ (from √ʔacus ‘face’ +

-šəd ‘lower leg’), and xʷʔəcusačiʔ ‘palm of hand’ (from √ʔacus ‘face’ + -ačiʔ ‘hand’). Most of

these contain a lexical suffix which usually requires dxʷ(s)- in other stems (see Section 2.1.1.1

above). As Hess (1995: 25 – 28) notes, xʷ- is almost certainly related diachronically to this

prefix, which expresses the general notion of internality or containment.

145

2.2.6 Seasonal pəd-

The prefix pəd- ‘seasonal [SEAS]’ is added to verbal and nominal bases to form nouns

referring to a season or time of year associated with a particular seasonal activity or event or with

a particular food item that becomes available at that time of year.

pədč’aʔəb ‘May’ (lit. ‘time of digging camas bulbs’)

(č’aʔəb ‘dig for roots’))

pədgʷədbixʷ ‘July’ (lit. ‘time of blackberries’)

(gʷədbixʷ ‘blackberry’)

pədhədəb ‘Summer’

(hədəb ‘be summer’ from √həd ‘be hot’)

pədkʷxʷic ‘September’ (lit. ‘time of silver salmon’)

(skʷxʷic ‘silver salmon’)

pədƛ’xʷayʔ ‘November’ (lit. ‘time of dog salmon’)

(ƛ’xʷayʔ ‘dog salmon’)

pədstəgʷad ‘June’ (lit. ‘time of salmonberries’)

(stəgʷad ‘salmonberry’)

pədt’aqa ‘August’ (lit. ‘time of salalberries’)

(t’aqa ‘time of salalberry’)

pədt’əs ‘Winter’

(º√t’əs ‘be cold (weather)’; cf. t’əsəb ‘be cold (weather)’)

pədxʷič’ib ‘October’ (lit. ‘time of elk/deer mating cry’)

(xʷič’ib ‘mating cry of elk/deer’)

pədxʷit’xʷit’il ‘October’ (lit. ‘time of falling’)

(xʷit’xʷit’il ‘large items falling’)

pədx&ʷiwaac ‘April’ (lit. ‘time of [robin] whistling’)

(x&ʷiwaac ‘whistle at ’)

Table 54: Nouns formed with pəd-

In addition to these, there are two words for ‘Spring’ which are not synchronically analyzable —

pədšiabac and pədƛ’qʷulil (possibly based on √ƛ’qʷ ‘emerge’). pəd- is also found prefixed to

the noun tab ‘thing’ to form the question word pədtab ‘when?’.

2.2.7 Colour x%i-

The prefix x&i- ‘colour’ is used to form nouns expressing basic colour terms from radicals for

colours:

x&ibəč’ ‘black’

(º√bəč’ ‘be black’) (NL)

x&ičəc ‘red’

(º√čc ‘be red’)

x&ikʷiƛ’ ‘red’ (Su)

(º√kʷiƛ’ ‘be red’)

x&ituc’ ‘black’

(º√tuc’ ‘be black’) (SL)

x&ʷiq’ʷixʷ ‘dark blue, dark green’

(º√q’ʷixʷ ‘be deep blue or dark green’)

x&ʷiqʷac ‘yellow, light green, pale’

(º√qʷac ‘yellow, light green, or pale’)

x&ʷiqʷəq’ʷ ‘white’

(º√qʷq’ʷ ‘be white’)

Table 55: Basic colour terms formed with x%i-

As can be seen from the forms in Table 55, the colour prefix has two phonologically-conditioned

allomorphs, [x&i-] and [x&ʷi-], the latter occurring with stems beginning in the rounded uvular

stops, /qʷ/ and /q’ʷ/. In addition to the five basic colour terms ‘black’, ‘white’, ‘red’,

This form is x&əkʷiƛ’ in Puyallup.

146

‘yellow/light green’, and ‘dark blue/dark green’, Lushootseed also has two additional colour

terms formed with x&i- — x&išukʷ ‘grey’ (from √šukʷ ‘powder’) and x&it’əc’ ‘green’ (from √t’əc’

‘bile’). Unlike the other colours, these two are formed from nouns.

Because the bases for the

forms in Table 55 are all bound radicals, it is difficult to assign them to a lexical class, although

they follow the derivational patterns of verbs and most likely belonged historically to that class.

2.2.8 Kinship prefixes ʔixʷ-, cixʷ-, and ciɬ-

There are three prefixes added to nominal bases to create words expressing various types of

kinship. These are ʔixʷ- ‘spouse’, cixʷ- ‘in-law’, and ciɬ- ‘half-sibling’.

The first of these prefixes, ʔixʷ- ‘spouse’, appears to be the most productive and occurs with

nominal bases of a variety of types including: age and kinship terms — ʔixʷč’ač’as ‘young

spouse’ (from č’ač’as ‘child’), ʔixʷčəgʷas ‘one who acts as wife’ [= ‘common-law spouse’?]

(from čəgʷas ‘wife’); professions — ʔixʷdxʷsp’ayəq ‘spouse of a canoe-maker’ (from

dxʷsp’ayəq ‘canoe-maker’), ʔixʷdxʷsčəbaʔ ‘spouse of a porter’ (from dxʷsčəbaʔ ‘porter’); and

numerical expressions — ʔixʷsəsaʔli ‘two spouses’ [= ‘pair of spouses’?] (from səsaʔli ‘two

people’).

The next prefix, cixʷ- ‘in-law’, is added only to kinship terms and is used to form words

expressing relationships by marriage in the same way that the English term in-law is used —

e.g., cixʷʔibac ‘grandchild’s spouse’ (from ʔibac ‘grandchild’), cixʷsyayaʔ ‘in-law’ (from yayaʔ

‘relative, friend’).

The more regular way of deriving colour terms from nominal radicals is to use the lexical suffix -alus ‘eye’. See

the discussion in Section *.*.

There is some uncertainty with respect to the part of speech of words formed with ʔixʷ- and, hence, to their

glosses. Some forms are glossed in the original sources as nouns while others are glossed as verbs, and the contexts

in which they appear are ambiguous as to lexical class. As Hess (1971) notes, there are no unambiguous attestations

of ʔixʷ- forms with possessive affixes, although this is attributed to an accidental gap in the corpus. ʔixʷ- forms are

treated here as (and re-glossed as) nouns, though the possibility remains that they are, in fact, verbs. The question

awaits the discovery of further contextualized attestations.

147

Finally, the prefix ciɬ- ‘half-sibling’ occurs in only two forms — ciɬbəsbad ‘half-sibling

with same father’ and ciɬbəsk’ʷuy ‘half-sibling with same mother’. Both of these are based on

possessive verbs stems — bəsbad ‘have a father’ and bəsk’ʷuy ‘have a mother’ — formed with

the propriative prefix bəs- (Section 2.1.5) from bad ‘father’ and sk’ʷuy ‘mother’, respectively.

2.2.9 Implement -təd

The suffix -təd ‘implement [IMPL]’ appears on a few stems, both nominal and verbal, to

create words expressing the instrument typically used in an action or associated with an object,

creating names for “items of traditional shape and established usage” (Hess 1971: 47). In a

number of these forms, -təd is reduced to [əd], while in others it appears as [ad].

These

reductions are somewhat idiosyncratic and lexicalized, and some forms (e.g., ɬik’ʷəd ‘gaff’, also

ɬik’ʷtəd) vary from speaker to speaker. A number of transparent -təd forms are given in Table

56:

c’aličtəd ‘umbrella’

(√c’al ‘be blocked from view’ + -ič ‘covering’)

cəq’disbad ‘fork’

(from º√cəq’ ‘be stuck into’ + -dis ‘tooth, tine’ + -b ‘middle’)

c’əstəd ‘nail’

(º√c’əs ‘be nailed’; cf. c’əsəd ‘nail ’)

čəbaʔtəd ‘tumpline’ (USk)

(√čəbaʔ ‘be loaded down with ’)

dᶻakʷtəd ‘rocking chair’

(√dᶻakʷ ‘be shaky, be shaking’)

liq’təd ‘red paint’

(º√liq’ ‘be painted red’; cf. liq’id ‘paint  red’)

ɬaqtəd ‘leg spindle’

(√ɬaq ‘spin wool’)

ɬik’ʷəd ‘gaff’

(√ɬik’ʷ ‘(to)fish’)

ƛ’akʷtəd ‘needle for cattail mats’

(º√ƛ’akʷ ‘be stitched’; cf. ƛ’agʷəd ‘stitch (mat)’)

ƛ’əgʷiʔčad ‘spear for bottom fish’

(√ƛ’agʷiʔč ‘spear bottom fish’)

p’ac’təd ‘sewing needle’

(√p’ac’ ‘sew’)

sx&usəbəd ‘straight razor’

(from √sax& ‘scrape’ + -us ‘face’ + -b ‘middle’)

šictəd ‘file’

(√šic ‘rub against ’)

t’ukʷtəd ‘ruler; tape measure’

(º√t’ukʷ ‘be measured’; cf. t’ugʷud ‘measure ’)

xʷc’agʷačid ‘bladder’

(from xʷ- ‘container’ + √c’akʷ ‘wash ’ + -ačiʔ ‘hand’)

xʷciqəbad ‘ramrod’

(from xʷ- ‘container’ + √ciq ‘be impaled’ + -b ‘middle’)

xʷiʔxʷiʔad ‘hunting gear’

(√xʷiʔxʷiʔ ‘hunt for , forage for ’)

xʷgʷədgʷatəd ‘voice; language’

(from xʷ- ‘container’ + √gʷad ‘talk’)

xʷqʷuʔəd ‘bucket’

(from xʷ- ‘container’ + √qʷuʔ ‘water’)

xʷsx&ʷaʔəd ‘bladder’

(from xʷ- ‘container’ + √sx&ʷaʔ ‘male urine’)

x&altəd ‘writing implement’

(√x&al ‘be written’)

yiq’ibad ‘awl for basket-making’

(from yiq’ib ‘make cedar-root baskets’)

There are some indications that the /a/ in the latter of the two allomorphs may have its origin in a separate

derivational affix /a/, as suggested in Hess (1967a: 34) and Hess & Hilbert (1976: vol. 2, 152–53). See Section 2.1.6

for further discussion.

148

Table 56: Nouns formed with -təd

In addition to these, there are less transparent words such as puʔtəd ‘shirt’ (possibly from √puʔ

‘be blown’), ƛ’aʔtəd ‘salmon weir’, k’ʷɬibəd ‘container made from thinly woven cedar slats’

(possibly from k’ʷəɬ ‘pour out, spill out’), šqači(ʔə)d ‘native-style hammer’ (possibly from šəq

‘be high’ + -ačiʔ ‘hand’), and čəc’disbad ‘toothpick’. Four of the forms in Table 56 also have

xʷ- ‘container’ (Section 2.2.6), as does the form xʷʔalʔaltəd ‘homeland’ (based on the noun

ʔalʔal ‘house’); unlike the others, this last form seems to have lost any sense of ‘implement’

from its meaning. The same is true of two kinship terms that appear to contain -təd: səxʷsqatəd

‘older siblings, older cousins’ (from √sqa ‘older sibling, older cousin’) and x&əɬtəd ‘man’s

brother-in-law’. The suffix -təd is textually infrequent and seems to be confined to a few highly

lexicalized forms. For further discussion of this issue in the context of the instrumental prefix

səxʷ-, see Section 2.2.3.

2.2.10 Relational -bid

The suffix -bid ‘relational [RLNL]’ is affixed most frequently to verbs and lexical adverbs to

form expressions of relative spatial relation, as in (109):

(109) a. ʔəskiisəxʷ ti hədli dhiwilbid

ʔəs–kiis=əxʷ ti hədli d–hiwil–bid

STAT–stand=now SPEC Henry 1SG.PO–go.ahead–RLNL

‘Henry is standing in front (lit. ‘ahead’) of me’

(Hess 1998: 43, ex. 1)

b. gʷəl ləgʷədiltub č’itbid ʔə tsiʔəʔ sɬadəyʔ

gʷəl lə=gʷədil–txʷ–b č’it–bid ʔə tsiʔəʔ sɬadəyʔ

SCONJ PROG=sit–ECS–PASS near–RLNL PR PROX:FEM woman

‘and they sat him near this woman’

(Hess 1998: 98, line 199)

In these expressions, -bid is used to create adverbials expressing the relative location of the

subject of the main predicate and the entity expressed as possessor of the -bid word.

Syntactically, these words pattern as adverbial adjuncts (*.*), not taking determiner and most

149

frequently appearing on the right edge of the clause. Also like adverbial adjuncts, they show the

potential for variable ordering, as can be seen by comparing the example in (109a) with that in

(110):

(110) ʔəskiisəxʷ dhiwilbid ti hədli

ʔəs–kiis=əxʷ d–hiwil–bid ti hədli

STAT–stand=now 1SG.PO–go.ahead–RLNL SPEC Henry

‘Henry is standing in front of me’

(Hess 1998: 43, ex. 4)

However, they also resemble relational nouns, both in that they have inherent possessors and that

they can appear in PPs and take determiners, as in (111):

(111) a. ʔəshuyucutəxʷ x%əɬ ti catcher ʔal tiʔiɬ ləqbid ʔə tiʔəʔ wiw’su

ʔəs–huyu–t–sut=əxʷ x%əɬ ti catcher ʔal tiʔiɬ ləq–bid

STAT–made–ICS–REFL=now seemingly catcher at DIST behind–RLNL

ʔə tiʔəʔ wiw’su

PR PROX children

‘he made himself sort of like a catcher there behind the children’

(lit. ‘he made himself sort of like a catcher at the children’s rear side’)

(Hess 1998: 98, line 293)

b. ʔal tiʔiɬ čaʔkʷbids əlgʷəʔ tusgʷəƛ’əlad ʔə tiʔiɬ

ʔal tiʔiɬ čaʔkʷ–bid–s əlgʷəʔ tu=s=gʷəƛ’əla–d ʔə tiʔiɬ

at DIST seaward–RLNL–3PO PL PAST=NM=stopped–ICS PR DIST

‘to the waterside of them he stops it (his arm)’

(lit. ‘that one’s stopping it is to their waterside’)

(Hess 2006: 55, line 311)

Words formed with -bid thus resemble dəxʷ=nominals (Section 7.4.2.2) in that they have

characteristics of both nouns and adverbials.

While expressions formed with -bid generally refer to relative spatial location, there are a

few examples in which -bid is extended metaphorically to express relative temporal location as

well, as in (112):

150

(112) x%aƛ’txʷ čəɬ kʷi gʷadst’ilib, gʷadsdᶻubalikʷ dᶻixʷbid ʔə kʷi ɬadsq’ʷəltubuɬ

x%aƛ’–txʷ čəɬ kʷi gʷə=ad=s=t’ilib gʷə=ad=s=dᶻubalikʷ

desire–ECS 1PL.SUB REM SBJ=2SG.PO=NM=sing SBJ=2SG.PO=NM=dance

dᶻix–bid ʔə kʷi ɬu=ad=s=q’ʷəl–t–ubuɬ

first–RNLN PR REM IRR=2SG.PO=NM=cooked–ICS–1PL.OBJ

‘we want you to sing and dance before you cook us’

[AJ Basket Ogress, line 84]

Hess (1998) also suggests that the metaphor of relative spatio-temporal location can also be

extended to relative location on a sliding scale of quality, as in (113):

(113) tiʔəʔ ʔəʔutx%s gʷəl hikʷbid ʔə tiʔəʔ sdəxʷiɬ

tiʔəʔ ʔəʔutx%s gʷəl hikʷ–bid ʔə tiʔəʔ sdəxʷiɬ

PROX Chinook.canoe SCONJ big–RLNL PR PROX hunting.canoe

‘as for the Chinook canoe, it is bigger than the hunting canoe’

(Hess 1998: 44, ex. 6)

As noted above in the discussion of comparative (Section 8.8), however, it is not clear if such

expressions with -bid are a productive method of comparative-formation, or if hikʷbid is a one-

off lexicalized form, this being the only stem found to date in this type of construction.

2.2.11 Other derivational affixes

In addition to the affixes listed above, there are a few noun-forming affixes that appear as

fossils in a limited number of words. These are given here for the sake of completeness, although

they do not appear to play any role in the synchronic grammar.

-bid ‘thing used for’

This suffix is found in four words with clearly identifiable bases: q’il’bid ‘canoe’ (from √q’il

‘be aboard’),

c’albid ‘shadow’ (from √c’al ‘be blocked from view’), qəlbid ‘waste, refuse’

(from √qəl ‘be bad’), and x&əčbid ‘intentions’ (from √x&əč ‘mind’). In addition, there are čsadbid

‘animal track’ (possibly from √čəs ‘be sent on an errand’), dxʷšadbid ‘imprint (on the ground)’

(possibly related to the lexical suffix -šad ‘lower leg’), x&acbid ‘food taken on a journey’

(possibly from √x&ac’ ‘cover something’), and x&ʷil’bid ‘cuckold’ (possibly from √x&ʷil’ ‘be lost, be

The glottalization of the /l/ in the form q’il’bid ‘canoe’ is an idiosyncratic effect of the -bid suffix.

David Beck 10-2-7 2:19 PM

Comment: -alc 'product' prdct

151

turned around’), as well as the completely opaque forms c’əx&bid ‘yew wood’, čəbid ‘Douglas

fir’, and q’əx&bid ‘hog fennel’.

hu- ‘vocative of endearment’

This prefix appears in two words, only one of which has an independently-attested base:

husk’ʷuy ‘dear little girl’ (from √sk’ʷuy ‘mother’) and hubaʔ ‘dear little boy’.

-aɬ ‘feminine diminutive’

This affix appears only in one common noun, sk’ʷuyaɬ (from √sk’ʷuy ‘mother’), a term of

endearment used for older women. However, it is also a common element in many female

names, a few of which have more or less clear derivational history (e.g., cisxʷisaɬ formed from

√xʷis ‘be brushed off’, a reference to a motion used by the person in her power dance).

2.3 Prepositions

Like many Salishan languages, Lushootseed has a limited number of words that can be

characterized as prepositions — that is, non-verbal elements that subcategorize for an NP

complement and which head adverbial and oblique argument phrases. In total there are five;

these are given in Table 57:

ʔə ‘of, by’

ʔal ‘at, on’

dxʷʔal ‘to, towards, into’

liɬʔal ‘via, through’

tul’ʔal ‘from, out of’

Table 57: Prepositions

These fall into two fairly obvious sets — the general preposition ʔ, a semantically bleached

element with primarily grammatical usesə (glossed here as ‘of, by’ more for convenience and

because of some functional overlap with the English of and by than as a reflection of its

meaning), and the lexically robust preposition ʔal and the words derived from this word using

the directional particles (Section 2.7.2). Prepositions of either type can take as a complement a

pronoun (114a), a bare noun (b), or a more complex noun phrase (c):

152

(114) a. ləɬčil čəd, capaʔ, dxʷʔal dəgʷi

lə=ɬčil čəd capaʔ dxʷ–ʔal dəgʷi

PRG–arrive 1SG.SUB grandfather CNTRPT–at you

‘I am coming, Grandfather, to you’

(Hess 2006: 28, line 157)

b. ʔux%ʷtubəxʷ tiʔiɬ sʔuladxʷ dxʷʔal tiʔiɬ sčətxʷəd

ʔux%ʷ–txʷ–b=əxʷ tiʔiɬ sʔuladxʷ dxʷ–ʔal tiʔiɬ sčətxʷəd

go–ECS–PASS=now DIST salmon CNTRPT–at DIST black.bear

‘the salmon was taken to Black Bear’

(Hess 1995: 154, line 67)

c. x%ʷul’əxʷ ʔiɬʔux%ʷ tsiʔəʔ diʔəʔ sk’ʷuys dxʷʔal tiʔəʔ haʔɬ šəgʷɬ

x%ʷul’=əxʷ ʔiɬ–ʔux%ʷ tsiʔəʔ diʔəʔ sk’ʷuy–s dxʷ–ʔal tiʔəʔ haʔɬ šəgʷɬ

only=now PRTV–go PROX:FEM here mother–3PO CNTRPT–at PROX good path

‘but his mother went on the good path anyway’

[AW Basket Ogress, line 39]

However, the complement of a preposition can also be headed by a member of any other

predicative class of words, including verbs and adverbs:

(115) a. ʔuʔukʷukʷ ʔə tə tib, gʷəl ləɬax%il

ʔu–ʔukʷukʷ ʔə tə tib gʷəl lə=ɬax%il

PFV–play PR SPEC be.strong SCONJ PROG=night

‘s/he played hard and evening came’

(Hess & Hilbert 1976: I, 50)

b. dəgʷi ɬadsugʷadgʷad dxʷʔal kʷi ɬuck’ʷaqid

dəgʷi ɬu=ad=s=ʔu–gʷadgʷad dxʷ–ʔal kʷi ɬu=ck’ʷaqid

you IRR=2SG.PO=NM=PFV–speak CNTRPT–at REM IRR=always

‘the one who will be speaking all the time will be you’

(Hess 2006: 8, line 139)

The complement of a preposition can also be, and very frequently is, a complex expression —

either a headless relative clause or a nominalization (Section 7.3), as in (116):

(116) a. huy, ʔabyidəxʷ ʔə tiʔiɬ x%ʷuˑl’ p’aƛ’aƛ’ stab

huy ʔab–yi–d=əxʷ ʔə tiʔiɬ x%ʷul’ p’aƛ’aƛ’ stab

SCONJ extend–DAT–ICS=now PR DIST only worthless what

‘and then he gave him just what was worthless’

(Hess 1995: 147, line 17)

David Beck 10-2-7 2:19 PM

Comment: the RC example is incorrect. In

fact, you need to reconsider the issue of verbal

complements of prepositions. Do a

concordance of dxw?al

153

b. x%ʷuˑƛ’udəxʷ əlgʷəʔ dxʷʔal kʷi gʷəsdᶻaq’s

x%ʷuƛ’u–d=əxʷ əlgʷəʔ dxʷ–ʔal kʷi gʷə=s=dᶻaq’=s

chewed–ICS=now PL CNTRPT–at REM SBJ=NM=fallen=3PO

‘they chewed on it so it would fall’

[AW Basket Ogress, line 110]

c. ləʔiʔilil ʔal tiʔəʔ dəxʷʔas

lə=ʔiʔilil ʔal tiʔəʔ dəxʷ=ʔa=s

PROG=making.noises at PROX ADNM=be.there=3PO

‘they were making noises where they were’

(Hess 2006: 76, line 808)

While the five prepositions are similar in terms of the types of complements they take, there are

some differences in the distributions and functions of the phrases they head. Phrases headed by

all five prepositions frequently serve as adverbial adjuncts to clauses (Section 8.2.7), and, with

the exception of ʔə, prepositions can also head predicate-phrases. On the other hand, ʔə (and not

the others) is used to introduce oblique objects (8.1.6) and agentive complements (8.1.7). ʔə also

differs from the others in not having much in the way of lexical meaning and functioning

primarily as a grammatical element. The meanings and uses of each of the five prepositions will

be discussed in more detail in the sections below.

2.3.1 Locative-temporal prepositions

The remaining four prepositions consist of the locative-temporal word ʔal and three others

formed by compounding this element with one of the directional particles — dxʷ, liɬ, and tul’

(Section 2.7.2). Like the Bella Coola (Nuxalk) system described by Nater (1984), the

Lushootseed system of prepositions is based on the notions of location and directed movement

relative to a deictic centre, as illustrated in Figure 2:

154

Figure 2: Lushootseed spatial prepositions

As indicated in this diagram, the basis of the system is the radical √ʔal which expresses location

at a point in space or time expressed by its complement:

(117) a. biƛ’itəbəxʷ ʔal tiʔiɬ čƛ’aʔ

biƛ’i–t–əb=əxʷ ʔal tiʔiɬ čƛ’aʔ

smashed–ICS–PASS=now at DIST rock

‘it was smashed on a rock’

(Hess 2006: 39, line 429)

b. huy bəgʷax%ʷ əlgʷəʔ ʔaləxʷ kʷi cədiɬ tushaydxʷsəxʷ əlgʷəʔ

huy bə=gʷax%ʷ əlgʷəʔ ʔal=əxʷ kʷi cədiɬ tu=s=hay–dxʷ=s=əxʷ

SCONJ ADD=walk PL at=now REM he PAST=NM=know–DC=3PO=now

əlgʷəʔ

‘then they walked on when they had it figured out’

(Hess 1998: 41, line 484)

The other prepositions build on this basic, static locative notion using directional particles. Thus,

the preposition dxʷʔal ‘toward’ combines the notions of location and centripetal motion, giving a

basic spatial meaning of motion towards a deictic centre or a temporal meaning expressing the

limit or termination of an event:

(118) a. gʷəl ʔux%ʷtubəxʷ dxʷʔal tiʔiɬ ʔalʔals

gʷəl ʔux%ʷ–txʷ–b=əxʷ dxʷ–ʔal tiʔiɬ ʔalʔal–s

SCONJ go–ECS–PASS=now CNTRPT–at DIST house–3PO

‘and so he

is taken [back] to his

house’

(Hess 1995: 153, line 63)

ʔal ‘locative’

dxʷʔal ‘centripetal’

liɬʔal ‘prolative’

tul’ʔal ‘centrifugal’

155

b. diˑɬ tushuy ʔə tiʔiɬ təkʷtəkʷəlus dxʷʔal tušac’s

diɬ tu=s=huy ʔə tiʔiɬ təkʷtəkʷəlus dxʷ–ʔal tu=s=šac’=s

FOC PAST=NM=be.done PR DIST owl CNTRPT–at PAST=NM=end=3PO

‘that is what Owl was doing to the end’

(Hess 2006: 10, line 177)

The preposition tul’ʔal, on the other hand, combines the locative base with the centrifugal

directional particle, and has the basic spatial meaning of motion away from a deictic centre and a

basic temporal to indicate the beginning of the period of time during which an event transpires:

(119) a. sət’cut tul’ʔal tiʔəʔ xʷʔax%ʷaʔad

sət’–t–sut tul’–ʔal tiʔəʔ xʷʔax%ʷaʔad

lifted–ICS–REFL CNTRFG–at PROX clam.basket

‘he lifts himself out of the clam basket’

[MS Basket Ogress, line 40]

b. tul’ʔal kʷi tushuy ʔə tiʔəʔ swatixʷəd sʔas ʔəsɬaɬlil

tul’–ʔal kʷi tu=s=huy ʔə tiʔəʔ swatixʷəd s=ʔa=s

CNTRFG–at REM PAST=NM=be.done PR PROX land NM=be.there=3PO

ʔəs–ɬaɬlil

STAT–live

‘from the making of this word she has been living there’

(Bierwert 1996: 185, line 60)

The fourth member of the system, liɬʔal, consists of the locative base and the prolative

directional particle, liɬ, and expresses location distributed throughout an area or motion within or

through a region:

(120) tiʔəʔ liɬʔal tiʔəʔ swatixʷtəd ƛ’udəxʷusaxʷəbs

tiʔəʔ liɬ–ʔal tiʔəʔ swatixʷtəd ƛ’u=dəxʷ=ʔu–saxʷəb=s

PROX PRLV–at PROX land HAB=ADNM=PFV–jump=3PO

‘here on the ground is where he runs’

(Hilbert & Hess 1977: 26)

Unlike the other prepositions, liɬʔal seems to have no temporal uses, nor does it have the range

of extended uses seen with the other spatial prepositions. The full set of meanings and contexts

for all four spatial prepositions will be dealt with individually in the sections below.

As illustrated in the preceding examples, Lushootseed spatial prepositions are used primarily

to introduce adjunct adverbial phrases (8.2.7). In addition to this function, however, prepositions

156

can also act as clausal predicates, in which case they are most naturally glossed as locative verbs

— ʔal ‘be at some place’, dxʷʔal ‘move towards some place’, tul’ʔal ‘move away from some

place, be from some place’, and liɬʔal ‘move through some area’). As shown by the examples in

(121), the resulting expressions tend to be rather convoluted from an English point of view and

are difficult to gloss naturally without obscuring the underlying syntactic structures:

(121) a. tux%ʷ tul’ʔal tə ʔa tucəxʷʔah cəxʷəsɬaɬlil

tux%ʷ tul’–ʔal tə ʔa tu=d=dəxʷ=ʔah d=dəxʷ=ʔəs–ɬaɬlil

only CNTRPT–at NSPEC be PAST=1SG.PO=ADNM=be 1SG.PO=ADNM=STAT–live

‘but over there is where I am from, [the place] where I live’

(lit. ‘it is just from the place I was, where I live’)

(Hess 1998: 79, line 46)

b. gʷəl, ʔal tiʔəʔ dbədaʔ tiʔəʔ ti adƛ’əwc’lax%ad, tiʔəʔ ti adɬidšəd, tiʔəʔ ti adstab, tiʔəʔ

adbitbitaʔs

gʷəl ʔal tiʔəʔ d–bədaʔ tiʔəʔ ti ad–ƛ’əwc’lax%ad tiʔəʔ ti

SCONJ at PROX 1SG.PO–child PROX SPEC 2SG.PO–arm.decorations PROX SPEC

ad–ɬidšəd tiʔəʔ ti ad–stab tiʔəʔ ad–bit–bitaʔs

2SG.PO–leg.tie PROX SPEC 2SG.PO–what PROX 2SG.PO–DSTR–breechcloth

‘so, here, my son, are your armbands, your leg-ties, your things, your breechcloths’

(lit. ‘so your armbands, leg-ties, etc., are at [you] my son’)

(Hess 2006: 35, line 322)

When used as locative predicates, prepositions take on some of the morphosyntactic properties of

verbs, such as the ability to be marked for aspect:

(122) a. siˑ lədxʷʔal tiʔəʔ dəxʷʔa(h) ʔə sp’ic’ikʷ …

si lə=dxʷ–ʔal tiʔəʔ dəxʷ=ʔah ʔə sp’ic’ikʷ

right.there PROG=CNTRPT–at PROX ADNM=be.there PR Diaper.Child

‘it was heading towards the place where Diaper Child was’

[DS Star Child, line 192]

b. diɬ k’ʷəɬ kʷədiʔ tubšədəd lədxʷʔal tsiʔəʔ bədaʔ ʔə tiʔəʔ tusbiaw ʔuʔatəbəd

diɬ k’ʷəɬ kʷədiʔ tubšədəd lə=dxʷʔal tsiʔəʔ bədaʔ ʔə tiʔəʔ

FOC QTV REM.DMA Sahaptin PROG=CNTRPT–at PROX:FEM child PR PROX

tu=sbiaw ʔu–ʔatəbəd

PAST=coyote PFV–die

‘it is that Sahaptin who is coming for the daughter of Coyote who has died’

(Hess 1998: 97, line 181)

157

In both of these examples, the preposition dxʷʔal bears the progressive aspect prefix lə-. No

other aspect is found associated with a preposition in the corpus, which may be an indication that

the ability of prepositions used as locative predicates to take aspect-markers is more limited than

that of true verbs. Note also that the context for the predicative use of dxʷʔal in (122b) is that of

a subject-centred relative clause, the predicative expression lədxʷʔal tsiʔəʔ bədaʔ ʔə tiʔəʔ

tusbiaw ‘[he] is coming for Coyote’s daughter’ modifying the noun, tubšədəd ‘Sahaptin’.

Prepositions used as locative predicates are attested other types of subordinate clauses as

well, as shown in (123):

(123) a. tux%ʷəxʷ stab gʷəl ʔaləs tadiʔ siq’gʷas ʔə tə šəgʷɬ

tux%ʷ=əxʷ stab gʷəl ʔal=əs tadiʔ s–siq’•gʷas ʔə tə šəgʷɬ

just=now what SCONJ PR=3SBJ DIST.DMA NP–spread•pair PR NSPEC path

‘they are just in that place were there is a fork in the road’

[AW Basket Ogress, line 99]

b. ʔəshaydxʷ čəd ti dəxʷtul’ʔaləp

ʔəs–hay–dxʷ čəd ti dəxʷ=tul’ʔal=ləp

STAT–know–DC 1SG.SUB SPEC ADNM=PR=2PL.PO

‘I know where you folks are from’

(Hess 2006: 68, line 628)

c. ɬux%ʷisid əlgʷəʔ bək’ʷ čad ʔal tiʔiɬ x%aƛ’ čad ʔəsɬəqʷdup dəxʷʔal tiʔiɬ ʔulʔal saʔdup

ɬu=x%ʷisid əlgʷəʔ bək’ʷ čad ʔal tiʔiɬ x%aƛ’ čad ʔəs–ɬəqʷ•dup

IRR=make.noise PL all where at DIST brush where STAT–wet•land

dəxʷ=ʔal=s tiʔiɬ ʔulʔal saʔ•dup

ADNM=at=3PO DIST bullrush bad•land

‘they will make noise everywhere, in the brush, where the land is wet, in places where

there are bulrushes, bad land’

(Hess 2006: 10, line 186)

(123a) the clause headed by the preposition ʔal is a subjunctive subordinate clause (Section 9.3)

and the preposition hosts the subjunctive subject clitic =əs. In the next two examples, the clause

headed by the prepositions is nominalized with the adjunct nominalizer dəxʷ= (Section 7.4.2.2),

creating abstract expressions of place. Not surprisingly, there are no attested instances of

158

prepositional predicates nominalized with s=, which is more frequently found in nominalizations

of the arguments of verbs or the nominalization of events than locations.

In context prepositional predicates can appear without an overt complement:

(124) a. dxʷʔaləxʷ ƛ’ubəxʷ [x%ʷul’əxʷ] ʔuhuyil təkʷtəkʷəlus

dxʷ–ʔal=əxʷ ƛ’ub=əxʷ x%ʷul’=əxʷ ʔu–huy–il təkʷtəkʷəlus

CNTRPT–at=now okay=now only=now PFV–be.done–INCH owl

‘it is for this reason, he had just better become an owl’

(Hess 2006: 8, line 154)

b. gʷəl tul’ʔal kʷi dᶻixʷ ʔaciɬtalbixʷ tuhuy ʔal tiʔəʔ swatixʷtəd …

gʷəl tul’–ʔal kʷi dᶻixʷ ʔaciɬtalbixʷ tu=huy ʔal tiʔəʔ swatixʷtəd

SCONJ CNTRFG–at REM first person PAST=be.done at PROX country

‘and the people who made this country came from [there] …’

[DS Star Child, line 2]

In expressions like these, the referent of the complement of the prepositional predicate (the

reasons for Owl’s transformation into a bird in 124a and the place that the first people came from

in 124b) are known from the surrounding discourse and their antecedents can be inferred from

context, allowing for the elision of any overt NP complement of the preposition. This

phenomenon is not attested with prepositions in their non-predicative uses.

As shown in the examples in (114) and (115), the complements of prepositions can be simple

and complex noun phrases, as well as non-verbal and clausal elements used as referential

expressions. As with most such expressions, prepositional complements are introduced by a

determiner. However, spatial prepositions in both their predicative and non-predicative uses are

attested occasionally with complements that lack an introductory determiner — and, in fact,

there are a number of environments in which this lack of a determiner seems to be fairly

common. One of these is with personal pronouns and proper names:

(125) a. gʷəl xʷiʔ ɬuləx%əb dxʷʔal dəgʷi

gʷəl xʷiʔ ɬu=lə=x%əb dxʷ–ʔal dəgʷi

INTJ not IRR=NEGP=be.heavy CNTRPT–at you

‘and it will not be heavy for you’

(Hess 1998: 81, line 92)

159

b. diɬəxʷ dəxʷt’uk’ʷəxʷ ʔə tiʔiʔəʔ stawixʷaʔɬ dxʷʔal xʷiwuʔc

diɬ=əxʷ dəxʷ=t’uk’ʷ=əxʷ ʔə tiʔ–iʔəʔ stawixʷaʔɬ dxʷ–ʔal xʷiwuʔc

FOC=now ADNM=go.home=now PR EXC–PROX children CNTRPT–at Xwiwuts

‘that is how the children got home to Xwiwuts’

[MS Basket Ogress, line 69]

b. t’ul’ʔahəxʷ gʷəl t’uk’ʷəxʷ dxʷʔal ʔəcəladiʔ

t’ul’–ʔa=həxʷ gʷəl t’uk’ʷ=əxʷ dxʷ–ʔal ʔəcəladiʔ

CNTRFG–be.there=now SCONJ go.home=now CNTRPT–at Utsallady

‘from there he went home to Utsallady’

[MS Basket Ogress, line 45]

The determiner is also quite frequently omitted with the noun ʔalʔal ‘house’ in contexts where

this noun refers to someone’s home:

(126) a. ɬčiˑldxʷ dxʷʔal ʔalʔals tiʔəʔ wiw’su

ɬčil–dxʷ dxʷ–ʔal ʔalʔal–s tiʔəʔ wiw’su

arrive–DC CNTRPT–at house–3PO PROX children

‘she managed to get the children to her house’

[AJ Basket Ogress, line 59]

b. ʔəsx%əɬ ʔal ʔalʔal

ʔəs–x%əɬ ʔal ʔalʔal

stat–sick at house

‘she was sick at home’

(Hess 2006: 12, line 41)

c. ʔal ʔalʔal tiʔəʔ bədaʔs …

ʔal ʔalʔal tiʔəʔ bədaʔ–s

at house PROX child–3PO

‘his son was at home’

(Hess 2006: 24, line 43)

d. x%ʷul’ čəxʷ ɬ[u]ɬaʔ dxʷʔal ʔalʔal čxʷa ɬux%ix%ilix%txʷ kʷi stawixʷəʔɬ

x%ʷul’ čəxʷ ɬu=ɬaʔ dxʷ–ʔal ʔalʔal čxʷa ɬu=x%i–x%ilix%–txʷ

only 2SG.SUB IRR=arrive CNTRPT–at house 2SG.COORD IRR=ATTN–fight–ECS

kʷi stawixʷəʔɬ

REM children

‘when you get home, you will compete with the children’

[MW Star Child, line 95]

As this example is presented in Hess (2006), it has an editorial amendment adding the specific determiner ti;

however, given the number of instances of the expression without the determiner in the corpus, it seems likely that

the utterance as it is recorded on tape is also correct.

160

This is reminiscent of similar phenomenon in languages like Spanish, which omits the definite

article in such expressions (voy a casa ‘I’m going home’). Another idiomatic expression which

is attested with a bare NP complement is ʔal x&əč (lit. ‘at mind’), which is used to attribute

emotional states:

(127) taaʔ haʔɬ ʔal x%əč ʔə tiʔəʔ caadiɬ

taaʔ haʔɬ ʔal x%əč ʔə tiʔəʔ caadiɬ

really good at mind PR PROX they

‘they are favorably impressed [by his reply]’

(Hess 1998: 80, line 64)

However, ʔal x&əč is also attested with a determiner in the NP complement in another context:

(128) x%ʷul’ ʔəsx%icil ʔal kʷi x%əčs

x%ʷul’ ʔəs–x%icil ʔal kʷi x%əč–s

only STAT–angry at REM mind–3PO

‘she was simply angry in her mind’

(Hess 2006: 31, line 236)

It could be that the presence of the possessive affix -s on the noun makes the complement

necessary referential (a particular person’s mind), or it could be that the idiom in (127) is

actually haʔɬ ʔal x&əč ‘be impressed’, and the sentence in (128) is more compositional.

As alluded in the discussion of the sentence in (128), another factor that seems to govern the

presence (and absence) of determiners in the complements of prepositions is the referentiality of

the expression — specifically, whether or not the PP serves to localize an event at a particular

place which is being singled out and localized for the hearer. In these cases, the complement is

introduced by determiner; otherwise, if the complement refers to a general region, direction, or a

non-specific entity, the determiner is omissible. For example, the PPs in the following examples

do not locate an event at a particular point or specified region in space, but instead localize it in a

more general direction relative the location of the event expressed by the matrix clause:

161

(129) a. gʷəl ʔix%ʷitəb dxʷčaʔkʷ dxʷʔal qʷuʔ

gʷəl ʔix%ʷi–t–əb dxʷ–čaʔkʷ dxʷ–ʔal qʷuʔ

SCONJ throw.away–ICS–PASS CNTRPT–seaward CNTRPT–at water

‘and he threw him down to the water’

(Hess 1998: 69, line 125)

b. x%əɬ ti dᶻəɬ tul’ʔal stuləkʷ kʷi dəxʷqʷicəxʷ əlgʷəʔ

x%əɬ ti dᶻəɬ tul’–ʔal stuləkʷ kʷi dəxʷ=qʷic=s=əxʷ əlgʷəʔ

seemingly PTCL CNTRFG–at river REM ADNM=downstream=3PO=now PL

‘it seems she came out of the river downstream from them’

[ML Basket Ogress, line 205]

c. ʔuqʷiʔad ʔal didiʔucid ʔaciɬtalbixʷ

ʔu–qʷiʔad ʔal di–diʔ•ucid ʔaciɬtalbixʷ

PFV–call.out at ATTN–opposite.side•mouth person

‘a person on the other side of the river is calling’

(Hess 1998: 96, line 151)

Similarly, the determiner can be omitted when a generalized or approximate location is being

expressed:

(130) ləhiqil dxʷʔal šqalicuts ʔal tiʔəʔ diʔəʔ hikʷ xʷʔax%ʷaʔads ʔəsčəbaʔtəb ʔə tsiʔəʔ diʔəʔ

lə=hiq–il dxʷ–ʔal s=šq–ali–t–sut=s ʔal tiʔəʔ

PROG=pushed–INCH CNTRPT–at NM=be.high–container–ICS–REFL=3PO at PROX

diʔəʔ hikʷ xʷʔax%ʷaʔad–s ʔəs–čəbaʔ–t–əb ʔə tsiʔəʔ diʔəʔ

here big clam.basket–3PO STAT–laden–ICS–PASS PR PROX:FEM he

‘he worked his way up to the top of the big clam basket she was carrying’

[DM Basket Ogress, line 21]

The same pattern is seen with temporal expressions: when a PP expresses a specific point in

time, the determiner is required, and when it expresses a more general notion of time of day or

the quantification of a period of time the determiner is omissible. Thus, temporal expressions

such as ʔal sɬax& ‘at night’ and in ʔal səɬax&il ‘in the evening’ are often attested as bare nominal

complements of prepositions used both as adjuncts (131a) and clausal predicates (131b):

Such expressions do not always lack determiners. Compare the sentences in (131) with (i):

(i) gʷəl diɬ tiʔəʔ diʔəʔ ɬukʷaɬ tə ləʔibəš ʔal tə ɬax%

gʷəl diɬ tiʔəʔ diʔəʔ ɬukʷaɬ tə ləʔibəš ʔal tə ɬax%

SCONJ FOC PROX he sun NSPEC PROG=travel PR NSPEC night

‘and the sun travels at night’

[MW Star Child, line 144]

162

(131) a. ʔu:, ƛ’a:l’ bəxʷəlušəd tiʔəʔ bəsʔabyitəbs ʔal bəsəɬax%il bəsʔuʔəɬəd ʔə tiʔəʔ wiw’su

ʔu ƛ’al’ bə=xʷəlušəd tiʔəʔ bə=s=ʔab–yi–t–əb=s ʔal

INTJ also ADD=fish.tail PROX ADD=NM=extend–DAT–ICS–PASS=3PO at

bə=səɬax%il bə=s=ʔu–ʔəɬəd ʔə tiʔəʔ wiw’su

ADD=evening ADD=NM=PFV–eat PR PROX children

‘oh, again what he was given to eat by the children in the evening [was] fish tail’

[AJ Basket Ogress, line 29]

b. ʔaləxʷ sɬax% kʷi tusʔəƛ’s tiʔəʔ shuysəxʷ sɬukʷalb

ʔal=əxʷ sɬax% kʷi tu=s=ʔəƛ’=s tiʔəʔ s=huy=s=əxʷ sɬukʷalb

at=now night REM PAST=NM=come=3PO PROX NM=be.done=3PO=now moon

‘the coming of he who became the moon [was] at night’

[HM Star Child, line 192]

Similarly, expressions of duration of this type may also lack determiners:

(132) ƛ’uʔux%ʷ p’ip’ic’ikʷ ʔal [də]č’axʷdat

ƛ’u=ʔux%ʷ p’ip’ic’ikʷ ʔal dəč’–axʷ•dat

HAB=go p’ip’ic’ikʷ at one–INCRP•day

‘Sp’ip’ic’ikʷ would go for one day’

[HM Star Child, line 134]

The same observation can be made about more complex temporal expressions: when used non-

referentially, these, too lack determiners. Compare (133a), which refers to a specific period of

time, to the non-referential temporal expressions in (133b):

(133) a. bək’ʷ sləx%il ʔal kʷi sluƛ’ils

bək’ʷ sləx%il ʔal kʷi s=luƛ’–il=s

all day at REM NM=elder–INCH=3PO

‘it was everyday as he grew [that he went hunting]’

[HM Star Child, line 79]

b. bəʔux%ʷ sləx%iləs tiʔəʔ sp’ip’ic’ikʷ ʔal sbuusaɬdalicuts

bə=ʔux%ʷ sləx%il=əs tiʔəʔ sp’ip’ic’ikʷ ʔal s=buusaɬdalicut=s

ADD=go day=3SBJ PROX Sp’ip’ic’ikʷ at NM=make.four.days=3PO

‘when it’s daylight, Sp’ip’ic’ikʷ goes again which is the fourth day for him’

[HM Star Child, line 146]

The complements of PPs are also quite consistently lacking in the temporal adjuncts to negative

clauses (Section 8.5), as in the following examples:

The difference between these expression might be compared to the difference between English expressions such as

at night versus in the night, although (as consideration of the English examples reveals) the precise nuances of the

distinction are difficult to glean from texts.

163

(134) a. xʷiʔəxʷ gʷəbəsx%ʷubils dxʷʔaləxʷ sbiɬaʔiləxʷ ʔə tiʔəʔ təkʷtəkʷəlus

xʷiʔ=əxʷ gʷə=bə=s=x%ʷubil=s dxʷ–ʔal=əxʷ s=biɬaʔil=əxʷ

NEG=now SBJ=ADD=NM=be.quiet=3PO CNTRPT–at=now NM=be.fed.up=now

ʔə tiʔəʔ təkʷtəkʷəlus

PR PROX owl

‘she [would] not keep quiet until [finally] Owl got fed up’

(Hess 2006: 4, line 44)

b. xʷiʔəxʷ gʷəsx%ʷubils dxʷʔal tucutəbsəxʷ ʔə tiʔəʔ cədiɬ dukʷibəɬ

xʷiʔ=əxʷ gʷə=s=x%ʷubil=s dxʷ–ʔal tu=s=cut–t–əb=s=əxʷ

NEG=now SBJ=NM=be.quiet=3PO CNTRPT–at PAST=NM=say–ICS–PASS=3PO=now

ʔə tiʔəʔ cədiɬ dukʷibəɬ

PR PROX he Changer

‘she did not shut up until Changer spoke’

(Hess 2006: 7, line 103)

c. ʔal suʔəɬəds ʔə tiʔiɬ sʔuladxʷs gʷəl xʷiˑʔ kʷi gʷəsbəkʷdxʷs

ʔal s=ʔu–ʔəɬəd=s ʔə tiʔiɬ sʔuladxʷ–s gʷəl xʷiʔ kʷi

at NM=PFV–eat=3PO PR DIST salmon–3PO SCONJ NEG REM

gʷə=s=bəkʷ–dxʷ=s

SBJ=NM=all–DC=3PO

‘as he was eating the salmon he could not finish it all’

(Hess 1998: 152, line 23)

The treatment of a temporal adjunct to an event which is considered unrealized at reference time

as non-referential seems logical; however, it should be noted that the events expressed in (134a)

and (b) (though not in 134c) are in fact realized after the time of reference (in fact, they are

expressed as the terminal temporal boundaries of the events expressed in the matrix clause).

Abstract uses of PPs also surface with complements that lack determiners. Here, too, the

conditioning factor seems to be one of referentiality versus non-referentiality. For instance,

complements of prepositions used to introduce expressions of general purpose or motive are

often found without determiners:

164

(135) a. xʷədᶻk’ʷcut tiʔəʔ skikəwič dxʷʔal sʔalils tiʔəʔ ʔiɬšəq

xʷədᶻk’ʷ–t–sut tiʔəʔ s–ki–kəwič dxʷ–ʔal s=ʔal–il=s

squirm–ICS–REFL PROX NP–ATTN–hunchback CNTRPT–at NM=at–INCH=3PO

tiʔəʔ ʔiɬ–šəq

PROX PRTV–be.high

‘Little Hunchback lifted himself up and out so that he would be even higher’

[AJ Basket Ogress, line 53]

b. hay, gʷəlaltəbəxʷ tsiʔəʔ sxʷəyuq’ʷ dxʷʔal sʔatəbəds

hay gʷəlal–t–əb–əxʷ tsiʔəʔ Basket.Ogress dxʷ–ʔal s=ʔatəbəd=s

SCONJ kill–ICS–PASS–now PROX:FEM Basket.Ogress CNTRPT–at NM=die=3PO

‘then they beat the Basket Ogress so that she died’

[ML Basket Ogress, line 122]

c. hay gʷəl, ƛ’alilcəxʷ, ƛ’alilcəxʷ ʔə tə č’ƛ’aʔ dxʷʔal sx%ʷuyilcəbsəxʷ ʔə tiʔəʔ č’ƛ’aʔ

hay gʷəl ƛ’al•ilc–əxʷ ƛ’al•ilc–əxʷ ʔə tə č’ƛ’aʔ dxʷʔal

SCONJ SCONJ cover•sphere=now cover•sphere=now PR NSPEC stone PR

s=x%ʷuy•ilc–əb=s=əxʷ ʔə tiʔəʔ č’ƛ’aʔ

NM=steam.cook•sphere–MD=3PO=now PR PROX stone

‘and then she covered [the fire] with stones so [they’d] be steam-cooked by the stones’

[AJ Basket Ogress, line 60]

d. hiˑgʷəxʷ stuləkʷəxʷ tiʔiɬ shuyitəbs əlgʷəʔ dxʷʔaləxʷ gʷəsp’əq’ʷs əlgʷəʔ

higʷ=əxʷ stuləkʷ=əxʷ tiʔiɬ s=huy–yi–t–əb=s əlgʷəʔ

big=now river=now DIST NM=be.done–DAT–ICS–PASS=3PO PL

dxʷ–ʔal=əxʷ gʷə=s=p’əq’ʷ=s əlgʷəʔ

CNTRPT–at=now SBJ=NM=float=3PO PL

‘a really big river was now made for them to float [across]’

(Hess 2006: 36, line 358)

It seems, in these cases, that the absence of the determiner is linked to the notion of an abstract

desired state of affairs or outcome, rather than the realization of a specific goal. This can be

contrasted with an example like (136), where the purpose of the action expressed by the main

clause is a very definite state of affairs:

(136) hay huyutəbəxʷ dxʷʔal kʷi gʷəsəsčəbaʔs

hay huyu–t–əb=əxʷ dxʷ–ʔal kʷi gʷə=s=ʔəs–čəbaʔ=s

SCONJ be.done–ICS–PASS=NOW CNTRPT–at REM SBJ=NM=STAT–laden=3PO

‘so, it was fixed up so that it could be backpacked’

(Hess 1998: 80, line 82)

165

The conditions for the item being prepared (in this case, an elk carcass) becoming portable in

(136) are fairly specific, whereas some of the situations in (135) are vaguer, and the conditions

under which they are fulfilled corresponding to a larger range of possible worlds. Of course, how

general or specific the desired outcome of an event is seen to be is a matter of construal on the

part of the speaker, and the use and non-use of determiners in the complements of prepositional

phrases in such constructions is by no means predictable in every case. In all likelihood, it

depends also on issues of style and usage that will require further research to unravel.

Locative prepositions are also found in a few contexts playing a role similar to that of

subordinating conjunctions in English. In these uses, the preposition introduces an adverbial

adjunct clause (Section *.*) in the form of a finite (non-nominalized) clause:

(137) a. ʔal bəƛ’uɬax% gʷəl bəʔux%ʷcəb ʔə bibščəb

ʔal bə=ƛ’u=ɬax% gʷəl bə=ʔux%ʷ–c–əb ʔə bi–bəščəb

at ADD=HAB=dark SCONJ ADD=go–ALTV–PASS PR ATTN–mink

‘when it would again be night, Little Mink would again go after it’

(Hilbert & Hess 1977: 24)

b. gʷəl lək’ʷilitəb əlgʷəʔ ʔal gʷəluutəb

gʷəl lə=k’ʷili–t–əb əlgʷəʔ ʔ al gʷə=lu–t–əb

SCONJ PROG=peek–ICS–PASS PL at SBJ=hear–ICS–PASS

‘and they were peered at as they were listened to’

(Hess 2006: 6, line 86)

Although these constructions resemble the determiner-less structures in (135), the subordinated

clauses here lack the nominalizing clitic s=. The difference between the nominalized and the

non-nominalized adjunct will be taken up in detail in Section *.*.

2.3.1.1 Locative ʔal ‘at, on’

When used as a preposition in its own right, ʔal indicates location at a point in space or time,

without reference to a origin or movement. Its most straightforward use is spatial, indicating

static location:

David Beck 10-2-7 2:19 PM

Comment: hypothesis: the finite subordinate

clauses reflect actual conditions that “contain”

the event in the matrix clause, the nominalized

are punctual, future, or hypothetical results.

166

(138) a. gʷəl (h)uy b[əʔ]əy’dxʷəxʷ tiʔəʔ qa tiʔiɬ sʔuladxʷ ʔal tiʔiɬ cədiɬ dəč’uʔ stuləkʷ

gʷəl huy bə=ʔəy’dxʷ=əxʷ tiʔəʔ qa tiʔiɬ sʔuladxʷ ʔal tiʔiɬ cədiɬ

SCONJ SCONJ ADD=find=now PROX many DIST salmon at DIST he

dəč’uʔ stuləkʷ

one river

‘and then he found a lot of salmon in this one river’

(Hess 1998: 66, line 23)

b. q’ʷəlbaxʷ tiʔəʔ cədiɬ bəščəb ʔal tiʔəʔ x%əɬ ti ɬudəxʷq’əlbs

q’ʷəl–b=axʷ tiʔəʔ cədiɬ bəščəb ʔal tiʔəʔ x%əɬ ti

ready.to.eat–MD=now PROX he mink at PROX seemingly

ɬu=dəxʷ=q’əlb=s

IRR=ADNM=camp=3PO

‘Mink roasted [it] at where he would sort of camp’

(Hess 1998: 66, line 27)

c. ləsdəkʷ ʔal tiʔəʔ čx%ʷəluʔ

ləs–dəkʷ ʔal tiʔəʔ čx%ʷəluʔ

PROG.STAT–inside at PROX whale

‘he is inside Whale (as Whale moves along)’

[ML Mink and Tutyika II, line 70]

In each of these examples, the preposition introduces a complement specifying a point in space

for the event expressed by the clause. Similarly, ʔal can be used to express location relative to a

person. In these it is amenable to the English gloss with:

(139) a. tastədᶻil ʔal tsiʔiɬ čəgʷas

tu=ʔas–tədᶻil ʔal tsiʔiɬ čəgʷas–s

PAST=stat–be.in.bed at DET:FEM wife–3PO

‘he

was in bed with his

wife’

(Hess 2006: 14, line 89)

b. ʔa čəɬ ʔal tiʔiɬ qʷi[qʷ]qʷistay’bixʷ

ʔa čəɬ ʔal tiʔiɬ qʷi–qʷ–qʷistay’bixʷ

exist 1PL.SUB at DIST ATTN–ATTN–dwarves

‘there we were with the dwarves’

(Hess 2006: 75, line 784)

In several instances, this seems to have been extended to introducing an adjunct phrase

expressing a beneficiary or interested party (a function more commonly associated with dxʷʔal):

167

(140) a. hay, tukʷukʷcutəxʷ ʔal tiʔiɬ syaʔyaʔs, sčətxʷəd

hay tu=kʷukʷcut=əxʷ ʔal tiʔiɬ syaʔyaʔ–s sčətxʷəd

SCONJ PAST=cook=now at DIST relative–3PO bear

‘next he cooked it for his relative, Black Bear’

(Hess 1995: 151, line 16)

b. gʷəl šaʔšaʔadad ʔal kʷi ʔəsqəp tsiʔiɬ cədiɬ sx%aʔhus

gʷəl šaʔšaʔadad ʔal kʷi ʔəs–qəp tsiʔiɬ cədiɬ sx%aʔhus

SCONJ leave.it.to at REM STAT–foolish DIST:FEM he sawbill

‘but leave it to someone as silly as Sawbill’

(Hess 2006: 32, line 249)

c. [yə]x%i huy ʔacʔaciɬtalbixʷ ʔal caadiɬ, tiʔəʔ caadiɬ ʔə qʷi[qʷ]qʷistay’bixʷ

yəx%i huy ʔac–ʔaciɬtalbixʷ ʔal caadiɬ tiʔəʔ caadiɬ ʔə qʷi–qʷi–qʷistay’bixʷ

because SCONJ DSTR–person at they PROX they PR ATTN–ATTN–dwarves

‘because they were people to them, [to] these who were dwarves’

(Hess 2006: 78, line 864)

The locative preposition is also occasionally found introducing the mention of an affected event-

participant in a semantic role (apparently) akin to that of PATIENT or UNDERGOER:

(141) haw’ ɬuʔəx%id ʔə til’x%i ti ʔal tiʔəʔ ʔuʔusil

haw’ ɬu=ʔəx%id ʔə til’x%i ti ʔal tiʔəʔ ʔu–ʔusil

INTJ IRR=what.happen PR later.on SPEC at PROX PFV–dive

‘oh, what is it that will happen presently to the one who dove?’

(Hess, 2006: 55, line 321)

It seems likely that the use of ʔal in this context is related to the fact that the event — and, hence,

the semantic role of the diver — is unknown, and so the event is described as taking place at the

diver’s location rather than specifying how exactly the diver is affected by or participates in the

event.

ʔal is also used to specify points or targets for an action:

(142) a. tiˑləb kʷi sʔas kʷi ƛ’uɬik’ʷ ʔal tiʔəʔ spčuʔs

tiləb kʷi s=ʔa=s kʷi ƛ’u=ɬik’ʷ ʔal tiʔəʔ spčuʔ–s

suddenly REM NM=be.there=3PO PROX HAB=hooked at PROX basket–3PO

‘there were abrupt tugs on her basket [as it got snagged]’

[AJ Basket Ogress, line 49]

David Beck 10-2-7 2:19 PM

Comment: English influence? see note in

reader

168

b. gʷəl lələč’, čəxʷə x%əqəd ʔal dəgʷi ʔə kʷi ƛ’əq’ʷ

gʷəl lə=ləč’ čəxʷə x%əq–əd ʔal dəgʷi ʔə kʷi ƛ’əq’ʷ

SCONJ ATTN–fill 2SG.COORD wrap–ICS at you PR REM watertight

‘fill it up a bit and you wrap it to yourself tightly’

[JS Basket Ogress, line 48]

In one case in the corpus, ʔal serves to introduce an object cognate with a lexical suffix, although

this is a role taken much more frequently by ʔə:

(143) tiləb ʔucaq’apsəbtəb ʔal tiʔiɬ scqapsəbs

tiləb ʔu–caq’•apsəb–t–əb ʔal tiʔiɬ scqapsəb–s

suddenly PFV–speared•throat–ICS–PASS at DIST neck–3PO

‘right away she was speared in the throat’

[ML Basket Ogress, line 106]

ʔal can also indicate that an action took place at a specific point within (as opposed to distributed

over) a wider area expressed by its complement:

(144) a. ʔuʔəy’ʔəy’dxʷ čəd tə sʔubədiʔ əlgʷəʔ ʔal kʷədiʔ t’aq’t čəda ʔuɬiltəb

ʔu–ʔəy’–ʔəy’dxʷ čəd tə sʔubədiʔ əlgʷəʔ ʔal kʷədiʔ t’aq’t

PFV–DSTR–find 1SG.SUB NSPEC hunter PL at REM.DMA inland

čəda ʔu–ɬil–t–əb

1SG.COORD PFV–give–ICS–PASS

‘I met some hunters up in the mountains and they gave it to me.

(Hess 1998: 82, line 127)

b. gʷəl huy ʔəɬədaxʷ əlgʷəʔ ʔal tudiʔ čəgʷalatxʷ

gʷəl huy ʔəɬəd=axʷ əlgʷəʔ ʔal tudiʔ čəgʷalatxʷ

SCONJ SCONJ eat=now PL at DIST.DMA outside.house

‘and then they ate over there outdoors’

(Hess 1998: 83, 140)

In at least one instance in the corpus, ʔal is used in an even looser sense, indicating location

relative to (but not located at or within) the region expressed by its complement:

(145) gʷəl tušudxʷ tiʔiɬ sʔuladxʷ ʔal tiʔiɬ sq’ax%ʷ

gʷəl tu=šuɬ–dxʷ tiʔiɬ sʔuladxʷ ʔal tiʔiɬ sq’ax%ʷ

SCONJ PAST=see–DC DIST salmon at DIST ice

‘and he saw a salmon through the ice’

(Hess 1995: 151, line 12)

Occasionally, ʔal surfaces in expressions of movement over an area, in which case it has an

inferred sense of distributed location:

169

(146) xʷuʔələʔ ʔa kʷi səʔibəš ʔal tiʔəʔ liɬʔilgʷil ʔə tə x%ʷəlč

xʷuʔələʔ ʔa kʷi səʔibəš ʔal tiʔəʔ liɬ–ʔilgʷil ʔə tə x%ʷəlc

maybe be.there REM NM=PROG=travel at PROX PRLV–shore PR NSPEC sea

‘I guess he was there traveling along the shore of the sea’

(Hess 1998: 65, line 15)

However, it is much more common to find liɬʔal used in such expressions.

Another frequent use of ʔal is to express the terminal point of a verb of motion:

(147) a. k’ʷit’əxʷ dxʷčaʔkʷ ʔal tiʔəʔ stuləkʷ

k’ʷit’=əxʷ dxʷčaʔ kʷ ʔal tiʔəʔ stuləkʷ

go.shoreward=now seaward at PROX river

‘he went down to the river’

(Hess 1995: 153, line 55)

b. saxʷəbid ʔal tiʔəʔ sq’ax%ʷ

saxʷəb–bi–d ʔal tiʔəʔ sq’ax%ʷ

jump–SS–ICS at PROX ice

‘he jumped after it [out on] to the ice’

(Hess 1995: 153, line 58)

The preposition here overlaps somewhat with dxʷʔal, although ʔal typically appears in

expressions where the emphasis of the speaker is less on the motion of the actor than on the

resultant location.

The notion of location at a point in space is naturally extended to temporal expressions of

actions taking place at a point in time:

(148) a. x%ʷul’ul’əxʷ p’q’ac ʔal kʷi tusɬčilsəxʷ

x%ʷul’–ul’=əxʷ p’q’ac ʔal kʷi tu=s=ɬčil=s=əxʷ

EXC–only=now rotten.wood at REM PAST=NM=arrive=3PO=now

‘it was nothing but a rotten log when he arrived’

(Hess 1998: 88, line 269)

b. gʷəl ləqaʔkʷ ʔal tiʔiɬ ƛ’ub dəxʷqaʔkʷs

gʷəl lə=qaʔkʷ ʔal tiʔiɬ ƛ’ub dəxʷ=qaʔkʷ=s

SCONJ PROG=rest at DIST well ADNM=rest=3PO

‘and he rests when it is good for him to rest [i.e., when he needs to]’

(Hess 1998: 82, line 114)

170

c. huy, ʔux%ʷtubəxʷ kʷi tust’əluʔb ʔal tiʔiɬ ʔəscil’cil

huy ʔux%ʷ–txʷ–b=əxʷ kʷi tu=st’əluʔb ʔal tiʔiɬ ʔəs–cil’–cil

SCONJ go–ECS–PASS=now REM PAST=dried.salmon at DIST STAT–DSTR–dished

‘then he took the dried king salmon as it was dished up’

(Hess 1998: 62, line 49)

d. xʷuʔələʔ čələp ɬasləqaxʷ ʔal kʷi ɬudxʷlaqəxʷ ʔə tiʔəʔ diʔəʔ

xʷuʔələʔ čələp ɬu=ʔas–ləq=axʷ ʔal kʷi ɬu=dxʷlaq=əxʷ

maybe 2PL.SUB IRR=STAT–listen=now at REM IRR=have.children=now

ʔə tiʔəʔ diʔəʔ

PR PROX here

‘perhaps you folks will listen when the parents speak’

[DM Basket Ogress, line 85]

These temporal expressions can refer to specific points in time (148a), repeated instances (b) and

(c), or future events (d). Although the temporal reference point for such expressions is most

naturally a verbal expression referring to an event, it is also possible for ʔal to introduce

expressions whose temporal reference point is a noun:

(149) xʷiʔ kʷi stab ɬusxʷiyuk’ʷ ʔal kʷi ɬulaq ɬuʔaciɬtalbixʷ ɬulək’ʷəd tiʔiɬ ʔaciɬtalbixʷ

xʷiʔ kʷi stab ɬu=sxʷiyuk’ʷ ʔal kʷi ɬu=laq ɬu=ʔaciɬtalbixʷ

NEG REM what IRR=Basket.Ogress at REM IRR=last IRR=person

ɬu=lək’ʷ–əd tiʔiɬ ʔaciɬtalbixʷ

IRR=eaten–ICS DIST people

‘there will be no Basket Ogresses in the time of the last people who will eat the people’

[JS Basket Ogress, line 71]

In this sentence, the time being referred to is the time of the laq ʔaciɬtalbixʷ ‘last people’, a time

at the very end of the era of myths before the world was prepared for human occupation by

Dukʷibəɬ, the Changer, the last (or, more accurately, penultimate) people being those alive at the

time of the transformation. The temporal reading of the PP ʔal kʷi ɬulaq ɬuʔaciɬtalbixʷ (lit. ‘at

the future last people’) is inferred from both the tense-marking and the pragmatics of the legend,

whose action is temporally removed from the ‘last people’.

Frequently, ʔal is also used to express an on-going event that is taking place while the event

expressed by the matrix clause occurs (cf. the expressions in 144 above):

David Beck 10-2-7 2:19 PM

Comment: weird construction

171

(150) a. ƛ’asluutəb ʔə tiʔəʔ suʔululuɬ ʔə tiʔəʔ bək’ʷ gʷat ʔal kʷi səɬax%il

ƛ’u=ʔas–luh–t–əb ʔə tiʔəʔ s=ʔu–ʔul–uluɬ ʔə tiʔəʔ

HAB=STAT–heard–ICS–PASS PR PROX NM=PFV–DIM.EFF–go.by.water PR PROX

bək’ʷ gʷat ʔal kʷi s=lə=ɬax%–il=s

all who at REM NM=PROG=dark–INCH=3PO

‘they are heard by everyone paddling about in the evening’

(Hess 2006: 13, line 59a)

b. gʷəl ʔəsaydubəxʷ ʔə tsiʔəʔ bədaʔs ʔal tiʔiɬ ƛ’usəsgʷədils ʔal kʷi [xʷ]cqʷuɬ

gʷəl ʔəs–hay–dxʷ–b=əxʷ ʔə tsiʔəʔ bədaʔ–s ʔal tiʔiɬ

SCONJ STAT–known–DC–PASS=now PR PROX:FEM child–3PO at DIST

ƛ’u=s=ʔəs–gʷədil=s ʔal kʷi xʷ–cqʷuɬ

HAB=NM=STAT–sit=3PO at REM CTD–day

‘and he was recognized by his daughter while he sat around in broad daylight’

(Hess 1998: 99, line 217)

c. bək’ʷ sləx%il ʔal kʷi sluƛ’ils

bək’ʷ s–ləx%–il ʔal kʷi s=luƛ’–il=s

all NM=light–INCH at REM NM=elder–INCH=3PO

‘everyday as he grew’

[HM Star Child, line 79]

There is also one instance in the corpus where ʔal is used to express a temporal limit (like the

English until):

(151) xʷiʔ gʷəsʔas kʷi sƛ’alqəb gʷux%ʷadᶻad kʷi wiw’su ʔal kʷi dxʷlaq

xʷiʔ gʷə=s=ʔa=s kʷi sƛ’alqəb gʷə=ʔu–x%ʷadᶻa–d kʷi

NEG SBJ=NM=exist=3PO REM monster SBJ=PFV–annihilate–ICS REM

wiw’su ʔal kʷi dxʷlaq

children at REM childless

‘there were no more monsters to have annihilated children until [they were] childless’

[ML Basket Ogress, line 230]

However, this type of expression more commonly uses dxʷʔal.

In addition to its spatial and temporal uses, ʔal is found in some more abstract contexts. One

of these is in apparent expressions of motive:

(152) a. haˑy gʷəl x%ix%ihəxʷ ʔal tiʔiɬ

hay gʷəl x%ix%ih=əxʷ ʔal tiʔiɬ

SCONJ SCONJ embarrassed=now at DIST

‘and so he [became] embarrassed about that’

(Hess 1998: 88, line 278)

172

b. hiiɬəxʷ əlgʷəʔ ʔal kʷədiʔ dəxʷʔa(h)s

hiiɬ=əxʷ əlgʷəʔ ʔal kʷədiʔ dəxʷ=ʔah=s

happy=now PL at REM.DMA ADNM=be.there=3PO

‘they

were happy for their

being there’

(Hess 2006: 63, line 517)

Generally, it is not clear in such expressions whether the notion of ‘motive’ is actually expressed

by the preposition or if it is a pragmatic implicature of, for example, a temporal reading — as,

for example, in 152a, which might literally mean ‘they were embarrassed then/at that point in

time’, or 152b, which might literally mean ‘they were happy while they were there/at their being

there’, the notion of ‘cause’ or ‘motive’ being implied by the context. These ambiguities in the

data underscore the difficulty of working on issues of semantics based largely on translation.

Other apparently abstract uses of ʔal include the occasional expression of manner,

(153) a. huy dəgʷatəbəxʷ tiʔəʔ sʔuləx% ʔal stab kʷi səshuytubs tiʔiɬ sʔuləx%

huy dəgʷa–t–əb=əxʷ tiʔəʔ sʔuləx% ʔal stab kʷi

SCONJ inside–ICS–PASS=now PROX dentalia at what REM

s=ʔəs–huy–txʷ–b=s tiʔiɬ sʔuləx%

NM=STAT–be.done–ECS–PASS=3PO DIST dentalia

‘then these dentalia were put in [Whale] however those dentalia were done up’

(Hess 2006: 67, line 606)

purpose,

(154) tuʔaʔ dᶻəɬ ti ƛ’əlayʔ gʷəl tux%ʷ xʷiʔ uʔxʷ gʷəsəsaxʷəbabacs ʔal kʷi sxʷəʔaʔxʷəʔils

tu=ʔaʔ dᶻəɬ ti ƛ’əlayʔ gʷəl tux%ʷ xʷiʔ uʔxʷ

PAST=be.there PTCL SPEC canoe SCONJ just NEG PTCL

gʷə=s=lə=saxʷəb•abac=s ʔal kʷi s=xʷəʔaʔxʷəʔ–il=s

SBJ=NM=PROG=jump•body=3PO at REM NM=swift–INCH=3PO

‘there must have been some shovel-nosed canoes but they could not yet jump over them

[in their practice] to become swift’

(Hilbert & Hess 1977: 14)

and topic of speech:

173

(155) hay gʷəl, tucutəxʷ tiʔəʔ qaw’qs ʔal tiʔəʔ sɬaq’əxʷ ʔə tiʔəʔ tatačulbixʷ

hay gʷəl tu=cut=əxʷ tiʔəʔ qaw’qs ʔal tiʔəʔ s=ɬaq’=əxʷ

SCONJ SCONJ PASt=speak=now PROX raven at PROX NM=fallen=now

ʔə tiʔəʔ tatačulbixʷ

PR PROX big.game.animal

‘then Raven spoke about the laying-out of the game animal’

[DS Star Child, line 297]

With verbs of speaking, what is said in more commonly introduced with ʔə, though the use of ʔal

in (155) may reflect that what is being reported is the topic of speech rather than the actual

contents of an utterance. Similarly, ʔal can be used to relate a particular individual to the topic of

a speaker’s utterance:

(156) ʔəsqil’il k’ʷiɬ … ʔal tiʔəʔ čxʷəluʔ

ʔəs–qil’il k’ʷiɬ ʔal tiʔəʔ čxʷəluʔ

STAT–lose.child QTV at PROX whale

‘they say that a child of Whale’s has died’ (i.e. ‘he lost a child they say about Whale’)

[ML Mink and Tutyika I, line 115]

In functions like these, ʔal seems to take on a largely grammaticized usage, that of establishing

an oblique relationship between an NP and the rest of the sentence. In this respect, ʔal resembles

the general preposition ʔə, although the syntactic relation marked by the former tends to be less

grammaticized and more oblique than the syntactic relation marked by the latter, which is more

often than not part of the government pattern of the verb.

Finally, in addition to being the morphological base for the other prepositions, ʔal is also the

base for a verb, ʔalil ‘come to some place, come to some time’, formed quite transparently from

the combination of the preposition and the inchoative suffix -il:

(157) a. ʔalil tiʔəʔ x%aləƛ’iqadis

ʔal–il tiʔəʔ x%aləƛ’iqadis

at–INCH PROX clearing

‘he came to a clearing’

(Bates, Hess & Hilbert 1994: 5)

174

b. gʷəl ʔalil tiʔiɬ bəsudᶻəlulč ʔə tsiʔəʔ diʔəʔ lədᶻubalikʷ

gʷəl ʔal–il tiʔiɬ bə=s=ʔu–dᶻəl•ulč ʔə tsiʔəʔ diʔəʔ

SCONJ at–INCH DIST ADD=NM=PFV–turn•belly PR PROX:FEM here

lə=dᶻubalikʷ

PROG=dance

‘then [the moment came that] she who was dancing turned her back [on them] again’

[DM Basket Ogress, line 64]

c. huy ʔaliləxʷ tiʔəʔ x%aʔx%alus

huy ʔal–il=əxʷ tiʔəʔ x%aʔx%alus

SCONJ at–INCH=now PROX raccoon

‘then they came to [i.e., turned their attention to] the raccoons’

(Bates, Hess & Hilbert 1994: 5)

In addition to the expected spatial and temporal uses shown in (157a) and (b), ʔalil has a more

figurative use, illustrated in (157c). It should be noted that only the temporal use in (157b) is

well-attested in the present analyzed corpus; the examples in (157a) and (c) are from a single

Skagit speaker of an older generation, which may mean these are archaic uses of the word. None

of the other prepositions are attested with -il or as the base for any other derivational affix.

2.3.1.2 Centripetal dxʷʔal ‘to, towards, into’

The preposition dxʷʔal is formed from the combination of ʔal with the centripetal directional

particle dxʷ (Section 2.7.2). Its basic meaning is to indicate motion or force directed towards a

point in space expressed by its complement. One of the most frequent uses of dxʷʔal is to

express the goal of a verb of motion:

(158) a. gʷəl ʔux%ʷtubəxʷ dxʷʔal tiʔiɬ ʔalʔals

gʷəl ʔux%ʷ–txʷ–b=əxʷ dxʷ–ʔal tiʔiɬ ʔalʔals

SCONJ go–ECS–PASS=now CNTRPT–at DIST house–3PO

‘and so he

took him

to his

house’

(Hess 1995: 153, line 63)

b. ʔux%ʷtubəxʷ tiʔiɬ sʔuladxʷ dxʷʔal tiʔiɬ sčətxʷəd

ʔux%ʷ–txʷ–b=əxʷ tiʔiɬ sʔuladxʷ dxʷ–ʔal tiʔiɬ sčətxʷəd

go–ECS–PASS=now DIST salmon CNTRPT–at DIST bear

‘he took the salmon to Black Bear’

(Hess 1995: 154, line 67)

175

As in (158a), the goal of motion can be a physical location or a person towards whom the action

is directed. The motion leading up to the location can be completed, as in (159),

(159) ƛ’uʔəƛ’ tiʔiɬə kikiʔalus stawixʷaʔɬ tubibƛ’aʔɬ dxʷʔal tiʔəʔ swədəbš

ƛ’u=ʔəƛ’ tiʔiɬə kikiʔalus stawixʷaʔɬ tu=bibƛ’aʔɬ dxʷ–ʔal tiʔəʔ

HAB=come DIST Kikiallus children PAST=have.picnic CNTRPT–at PROX

swədəbš

Swinomish

‘the Kikiallus children would come and have a picnic at Swinomish’

[MS Basket Ogress, line 1]

b. tiləb ʔudxʷcaq’ax%aditəb dxʷʔal tiʔiɬ xʷʔilax%ads

tiləb ʔu–dxʷ–caq’•ax%ad–di–t–əb dxʷ–ʔal tiʔiɬ xʷ–ʔilax%ad–s

suddenly PFV–CTD–spear•side–SS–ICS–PASS CNTRPT–at DIST CNTNR–side–3PO

‘right then he was speared in the side’

(Hess 2006: 15, line 96)

or incomplete, as in (160), where dxʷʔal indicates motion towards or in the direction of a

particular location:

(160) a. ʔux%ʷəxʷ dxʷʔal kʷədiʔ dəxʷɬaʔs əlgʷəʔ

ʔux%ʷ=əxʷ dxʷ–ʔal kʷədiʔ dəxʷ–ɬaʔ–s əlgʷəʔ

go=now CNTRPT–at REM.DMA ADNM=arrive=3PO PL

‘they went towards their destination’

[LA Basket Ogress, line 109]

b. x%ʷul’ ɬuč’itil dxʷʔal tiʔəʔ diʔəʔ hud ɬudᶻubalikʷəs

x%ʷul’ ɬu=č’it–il dxʷ–ʔal tiʔəʔ diʔəʔ hud ɬu=dᶻub–alikʷ=əs

only IRR=near–INCH CNTRPT–at PROX here fire IRR=kick–ACT=3SBJ

‘she will draw near the fire when she dances’

[DM Basket Ogress, line 59]

In some cases, dxʷʔal is used where there is not actual motion involved, but rather force is

directed towards the deictic centre:

(161) gʷəl ck’ʷaqidəxʷ əlgʷəʔ ʔəsbiq’id dxʷʔal tiʔəʔ hud

gʷəl ck’ʷaqid=əxʷ əlgʷəʔ ʔəs–biq’i–d dxʷ–ʔal tiʔəʔ hud

SCONJ always=now PL STAT–pressed–ICS CNTRPT–at PROX fire

‘then they kept pressing her down into the fire’

[MS Basket Ogress, line 61]

176

The motion implicit in the meaning of dxʷʔal can also be somewhat more metaphorical, as in the

case of dxʷʔal introducing the hearers with verbs of speaking:

(162) a. gʷəl yəcəbaxʷ dxʷʔal tiʔiɬ ʔiišəds

gʷəl yəc–əb=axʷ dxʷ–ʔal tiʔiɬ ʔiišəd–s

SCONJ tell–MD=now CNTRPT–at DIST relative–3PO

‘and he told his family’

(Hess 2006: 72, line 716)

b. təbəwihəxʷ sgʷəgʷadads dxʷʔal tiʔəʔ biʔbədaʔs

təbəwih=əxʷ s=gʷəgʷadad=s dxʷ–ʔal tiʔəʔ biʔ–bədaʔ–s

eneretically=now NM=talk=3PO CNTRPT–at PROX ATTN–child–3PO

‘she really went at it shushing her little baby’

(Hess 2006: 4, line 37)

This pattern is generally only found with intransitive verbs of speech; transitive verbs of speech

generally take the hearer as a direct object.

Because the central meaning of dxʷʔal is that of motion towards a deictic centre, it covers a

broad range of situation types that are treated differently, and which require different

prepositions, in languages like English. Lushootseed, for instance, does not explicitly distinguish

motion leading up to a point from motion leading to contact with the upper surface of an object

or area (English onto), both cases being covered by dxʷʔal:

(163) a. xʷit’iləxʷ dxʷʔal tudiʔ cədiɬ sq’ʷas(tə)dulic’aʔ

xʷit’il=əxʷ dxʷ–ʔal tudiʔ cədiɬ sq’ʷastədulic’aʔ

fall=now CNTRPT–at DIST.DMA he blanket

‘he dropped her way [down] on to that dog-and-goat-hair-blanket’

(Hess 2006: 20, line 201)

b. ʔuˑ tux%ədtəbaxʷəs ʔə tiʔəʔ ʔiɬluƛ’luƛ’ wiw’su tiˑləb dxʷʔal tiʔəʔ ʔəsx%ʷuyilcəb

č’əƛ’č’ƛ’aʔ

ʔu tu=x%əd–t–əb=axʷ=əs ʔə tiʔəʔ ʔiɬ–luƛ’–luƛ’ wiw’su

INTJ PASt=pushed–ICS–PASS=now=3SUB PR PROX PRTV–DSTR–old children

tiləb dxʷ–ʔal tiʔəʔ ʔəs–x%ʷuy•ilc–əb č’əƛ’–č’ƛ’aʔ

suddenly CNTRPT–at PROX STAT–steam.cook•round.object–MD DSTR–stone

‘oh, she was pushed by the children onto the stones for steam cooking’

[AJ Basket Ogress, line 100]

177

Nor does the language distinguish motion or force directed towards the interior of an object or

terminating in containment (English into):

(164) a. lədəgʷaš dxʷʔal tiʔəʔ xʷʔax%ʷaʔads

lə=dəgʷa–š dxʷ–ʔal tiʔəʔ xʷʔax%ʷaʔad–s

PROG=be.inside–ICS CNTRPT–at PROX clam.basket–3PO

‘she is putting them into her clam basket’

[MS Basket Ogress, line 13]

b. huy, dᶻəlqcutəxʷ tiʔiɬ č’ƛ’aʔ dxʷʔal tiʔiɬ [s]əsluʔ

huy dᶻəlq–t–sut=əxʷ tiʔiɬ č’ƛ’aʔ dxʷ–ʔal tiʔiɬ səsluʔ

SCONJ PFV–turn–ICS–REFL=now DIST stone CNTRPT–at DIST cave

‘then Rock turns himself around [covering the entrance] to the cave’

(Hess 1995: 149, line 49)

c. q’ilid əlgʷəʔ dxʷʔal tiʔəʔ q’il’bids

q’ili–d əlgʷəʔ dxʷ–ʔal tiʔəʔ q’il’bid–s

board–ICS PL CNTRPT–at PROX canoe–3PO

‘they put it into their [own] canoe’

(Hess 2006: 57, line 363)

Not infrequently, examples parallel to those in (164) are found with ʔal rather than dxʷʔal:

(165) a. x%ʷul’ ʔəsq’il ʔal tiʔəʔ q’il’bid

x%ʷul’ ʔəs–q’il ʔal tiʔəʔ q’il’bid

only STAT–be.aboard at PROX canoe

‘it was just on board in a canoe’

(Hess 1998: 92, line 38)

b. gʷəl lədəgʷəš ʔal tə syalts

gʷəl lə=dəgʷ–əš ʔal tə syalt–s

SCONJ PROG=inside–ICS at NSPEC basket–3PO

‘then she was putting them into her basket’

[JS Basket Ogress, line 51]

As noted earlier, the contrast between sentences that use ʔal and those that use dxʷʔal is likely to

be one of emphasis, the former emphasizing the final position of the object being localized and

the later the process by which the object arrives.

As with ʔal, dxʷʔal extends itself into the temporal realm. Its basic use is to express the limit

or termination of an event:

178

(166) a. ləsaxʷəbabac ʔə tiʔəʔ qʷɬaʔy dxʷʔal tiʔiɬ sɬčis tiʔəʔ ʔaʔyəds qaw’qs

lə=saxʷəb•abac ʔə tiʔəʔ qʷɬaʔy dxʷ–ʔal tiʔiɬ s=ɬčil=s tiʔəʔ

PROG=jump•body PR PROX log CNTRPT–at DIST NM=arrive=3PO PROX

ʔaʔyəd–s qaw’qs

friend–3PO raven

‘he is jumping over the logs until he arrives at his friend, Raven’

(Hilbert & Hess 1977: 25–26)

b. dxʷčad kʷi sugʷadx%ʷs əlgʷəʔ dxʷʔal tiʔiɬ tusq’əlbs əlgʷəʔ

dxʷ–čad kʷi s=ʔu–gʷadx%ʷ=s əlgʷəʔ dxʷ–ʔal tiʔiɬ

CNTRPT–where REM NM=PFV–walk=3PO PL CNTRPT–at DIST

tu=s=q’əlb=s əlgʷəʔ

PAST=NM=camp=3PO PL

‘they walked around until they camped’

(Hess 2006: 32, line 258)

In several cases, the preposition indicates that a state of affairs is extant at the time expressed by

its complement, without necessarily indicating that the state of affairs comes to an end at that

point in time:

(167) xʷiʔəxʷ gʷəsx%aabs dxʷʔal sɬčil ʔə tsiʔəʔ bədaʔs

xʷiʔ=əxʷ gʷə=s=x%aab=s dxʷ–ʔal s=ɬčil ʔə tsiʔəʔ bədaʔ–s

NEG=now SBJ=NM=cry=3PO CNTRPT–at NM=arrive PR PROX:FEM offspring–3PO

‘he isn’t crying when her [the old woman’s] daughter arrives’

[HM Star Child, line 48]

These uses seem fairly consistent with the basic meaning of the preposition, motion (in this case,

metaphorical motion through time) towards a particular spatial, or temporal, location. A little less

transparently, dxʷʔal is also used in the odd expression of duration, such as that in (168):

(168) xʷiʔəxʷ kʷi stabəxʷ gʷəƛ’ašudxʷ əlgʷəʔ dxʷʔaˑl kʷədiʔ tuk’ʷidəɬdat

xʷiʔ=əxʷ kʷi stab=əxʷ gʷə=ƛ’u=ʔas–šuɬ–dxʷ əlgʷəʔ dxʷ–ʔal kʷədiʔ

NEG=now REM what=now SBJ=HAB=PFV–see–DC PL CNTRPT–at REM.DMA

tu=k’ʷid•əɬ•dat

PAST=how.many•times•day

‘they could not see things for many days’

(Hess 2006: 53, line 263)

The word qaw’qs is given in the original source as kaw’qs; Hess (p.c.) now considers this an error.

179

It may be in these cases the translation is a little misleading, a more accurate gloss being ‘they

could not see anything until many days had passed’, bringing it into line with the examples in

(166); further research is needed on the issue.

Of all the spatial prepositions, dxʷʔal seems to have the widest variety of metaphorical or

non-basic uses, though most of these seem to be fairly logical extensions from its basic, literal

meaning. One such notion is that of goal, in particular the goal of a search, which bears an

obvious semantic similarity to the notion of motion towards a location:

(169) a. gʷič’gʷič’əxʷ dxʷʔal kʷi ɬudəxʷʔux%ʷs, dxʷčadəxʷ kʷi ɬuspaq’acuts tiʔəʔ ʔaciɬtalbixʷ

gʷič’–gʷič’=əxʷ dxʷ–ʔal kʷi ɬu=dəxʷ=ʔux%ʷ=s dxʷ–čad=əxʷ kʷi

DSTR–search=now CNTRPT–at REM IRR=NM=go=3PO CTRPT–where=now REM

ɬu=s=paq’a–t–sut=s tiʔəʔ ʔaciɬtalbixʷ

IRR=NM=distribute–ICS–REFL=3PO PROX person

‘they looked for where they could go, where the people could [re]settle’

(Hess 1998: 102, line 286)

b. bədiɬ kʷi ləbəsčəgʷas dxʷʔal tsiʔacəc bədaʔ ʔə tiʔiɬ siʔab ʔuʔatəbəd

bə=diɬ kʷi lə=bəs–čəgʷas dxʷ–ʔal tsiʔacəc bədaʔ ʔə tiʔiɬ

ADD=IDN REM PRG–have–wife CNTRPT–at UNQ:DEM:FEM child PR DIST

siʔab ʔu–ʔatəbəd

noble PFV–die

‘he is the one seeking as a wife this daughter of the nobleman who has died’

(Hess 1998: 98, line 189)

By the same token, the purposes for which actions are performed can be construed as

metaphorical goals, leading to uses of dxʷʔal in expressions such as those in (170):

(170) a. hay huyutəbəxʷ dxʷʔal kʷi gʷəsəsčəbaʔs

hay huyu–t–əb=əxʷ dxʷ–ʔal kʷi gʷə=s=ʔəs–čəbaʔ=s

SCONJ be.done–ICS–PASS=now CNTRPT–at REM SBJ=NM=STAT–laden=3PO

‘so, it was fixed up so that it could be backpacked’

(Hess 1998: 80, line 82)

180

b. tiʔiɬ tudəxʷhuyutəbsəxʷ dxʷʔal kʷi gʷədəxʷdukʷutəbsəxʷ

tiʔiɬ tu=dəxʷ=huyu–t–əb=s=əxʷ dxʷ–ʔal kʷi

DIST PAST=ADNM=be.done–ICS–PASS–3PO=now CNTRPT–at REM

gʷə=dəxʷ=dukʷu–t–əb=s=əxʷ

SBJ=ADNM=bewitch–ICS–PASS=3PO=now

‘that is why they did to him [as they did] in order to put a spell on him’

(Hess 1998: 101, line 264)

The same logic may apply to examples like that in (171) as well, in which dxʷʔal introduces a

command:

(171) ʔuhilitəb tə stawixʷaʔɬ dxʷʔal kʷi sʔux%ʷs dxʷʔal kʷi ʔilgʷiɬ ʔə kʷi x%ʷəlč

ʔu–hili–t–əb tə stawixʷaʔɬ dxʷ–ʔal kʷi s=ʔux%ʷ=s dxʷ–ʔal

PFV–order–ICS–PASS NSPEC children CNTRPT–at REM NM=go=3PO CNTRPT–at

kʷi ʔilgʷiɬ ʔə kʷi x%ʷəlč

REM shoreline PR REM sea

‘the children were told to go down to the shores of the sea’

[LA Basket Ogress, line 1]

However, this is the only such example in the present corpus, and jussives are more frequently

realized with other constructions (see Section *.*).

In addition to these uses, dxʷʔal appears in a few other contexts that seem somewhat less

transparently related to its basic meaning. One of these is in constructions that can be glossed

with the English prepositions for or with respect to:

(172) a. gʷəl xʷiʔ ɬuləx%əb dxʷʔal dəgʷi

gʷəl xʷiʔ ɬu=lə=x%əb dxʷ–ʔal dəgʷi

SCONJ NEG IRR=NEGP=heavy CNTRPT–at you

‘and it will not be heavy for you’

(Hess 1998: 81, line 92)

b. … tsi sxʷəyuq’ʷ, sƛ’alqəb dxʷʔal kʷi wiw’su

tsi sxʷəyuq’ʷ sƛ’alqəb dxʷ–ʔal kʷi wiw’su

SPEC:FEM Basket.Ogress monster CNTRPT–at REM children

‘… Basket Ogress, monster to the children’

[ML Basket Ogress, line 109]

181

c. ƛ’al’ b(ə)asʔistaʔ dxʷʔal tiʔəʔ təbɬ ʔi tiʔiɬ dəxʷudxʷliq’ʷusəbs …

ƛ’al’ bə=ʔas–ʔistaʔ dxʷ–ʔal tiʔəʔ təbɬ ʔi tiʔiɬ

also ADD=STAT–same CNTRPT–at DIST ochre CONJ DIST

dəxʷ=ʔu–dxʷ–liq’ʷ•us–əb=s

ADNM=PFV–CTD–paint•face–MD=3PO

‘it is also the same regarding the ochre and that used for painting one’s face …’

(Hess 1998: 92, line 32)

dxʷʔal is also found in some cases relating the topic of a speech act (cf. the examples with ʔal in

155 above):

(173) a. ʔəscuucəxʷ čəɬ tiʔiɬ čxʷəluʔ dxʷʔal gʷəlapu

ʔəs–cut•uc=əxʷ čəɬ tiʔiɬ čxʷəluʔ dxʷ–ʔal gʷəlapu

STAT–say•mouth=now 1PL.SUB DIST whale CNTRPT–at you.guys

‘we have told Whale about you folks’

(Hess 2006: 67, line 600)

b. … čɬa ɬuɬčiltxʷəxʷ dxʷʔal kʷi tushuyutəb čəɬ sʔušəbabdxʷ …

čɬa ɬu=ɬčil–txʷ=əxʷ dxʷ–ʔal kʷi tu=s=huyu–t–əb

1PL.COORD IRR=arrive–ECS=now CNTRPT–at REM PAST=NM=be.done–ICS–PASS

čəɬ sʔušəbabdxʷ

1PL.PO unfortunate.one

‘… and we will arrive with [the story] about how we were made unfortunates …’

(Hess 2006: 74, line 759)

Similarly, dxʷʔal appears in a few contexts where it seems to relate a motive for an event, action,

or state of mind:

(174) a. cick’ʷəxʷ ʔudxʷ(ʔ)idawligʷəd dxʷʔal tsiʔiɬ čəgʷas, x%ʷuʔx%ʷəyʔ

cick’ʷ=əxʷ ʔu–dxʷʔidawligʷəd dxʷ–ʔal tsiʔiɬ čəgʷas–s x%ʷuʔx%ʷəyʔ

very=now PFV–worry CNTRPT–at DIST:FEM wife–3PO Little.Diver

‘he is really worried about his wife’

(Hess 2006: 13, line 64)

b. lədxʷsx%ʷal’dxʷ[əb] dxʷʔal tiʔəʔ tusqadaditəbs tiʔəʔ tusʔuladxʷs

lə=dxʷs–x%ʷal’–dxʷ–əb dxʷ–ʔal tiʔəʔ tu=s=qada–di–t–əb=s

PRG–CTD–fail–DC–DSD CNTRPT–at PROX PAST=NM=steal–SS–ICS–PASS=3PO

tiʔəʔ tu=sʔuladxʷ–s

PROX PAST=salmon–3PO

‘he was wanting to get the best of him because he had stolen his salmon’

(Hess 1998: 69, line 118)

182

c. tugʷəlaltəb dxʷʔal tudəxʷukʷaxʷdubuts

tu=gʷəlal–t–əb dxʷ–ʔal tu=dəxʷ=ʔu–kʷaxʷ–dxʷ–but=s

PAST=punish–ICS–PASS CNTRPT–at PAST=ADNM=PFV–manage–DC–REFL=3PO

‘she had been beaten up because she had helped herself’

(Hess 2006: 21, line 223)

These uses of the preposition are the most difficult to relate to dxʷʔal’s basic meaning.

A final use of dxʷʔal is in the expression of comparisons:

(175) ʔiɬqʷiq’ʷ čəd dxʷʔal dəgʷi

ʔiɬ–qʷiq’ʷ čəd dxʷ–ʔal dəgʷi

PRTV–strong 1SG.SUB CNTRPT–at you

‘I am stronger than you’

(Bates, Hess & Hilbert 1994: 5)

Comparative constructions will be discussed further in Section 8.8.

2.3.1.3 Centrifugal tul’ʔal ‘from, out of’

The preposition tul’ʔal is formed from the combination of ʔal with the centrifugal directional

particle tul’ (Section 2.7.2). Its basic meaning is to indicate motion or force directed away from a

point in space expressed by its complement. This meaning is most explicit in expressions that

specify motion away from a point of origin:

(176) taxʷčəɬəb sʔəɬəd ʔə tiʔəʔ diʔəʔ stawixʷaʔɬ tasčəbaʔəd tul’ʔal tudiʔ čaʔkʷ

tu=ʔas–dxʷ–čəɬ–əb sʔəɬəd ʔə tiʔəʔ diʔəʔ stawixʷaʔɬ

PAST=STAT–CTD–make–DSD food PR PROX here children

tu=ʔas–čəbaʔ–əd tul’–ʔal tudiʔ čaʔkʷ

PAST=STAT–laden–ICS CNTRFG–at DIST.DMA waterward

‘she wanted to make food of the children she’d carried up from over there by the water’

[DM Basket Ogress, line 73]

tul’ʔal is also used to express motion coming out of an enclosed or containing space,

(177) a. ƛ’iqagʷiləxʷ tiʔiɬ sčətxʷəd tul’ʔal tiʔiɬ ʔalʔals

ƛ’iq–agʷil=əxʷ tiʔiɬ sčətxʷəd tul’–ʔal tiʔiɬ ʔalʔal–s

emerge–AUTO=now DIST bear CNTRFG–at DET house–3PO

‘Black Bear comes out of his house’

(Hess 1995: 143, line 6)

183

b. x%əɬ ti dᶻəɬ tul’ʔal stuləkʷ kʷi dəxʷqʷicəxʷ əlgʷəʔ

x%əɬ ti dᶻəɬ tul’–ʔal stuləkʷ kʷi dəxʷ=qʷic=s=əxʷ əlgʷəʔ

seem PTCL CNTRFG–at RIVER REM ADNM=downstream=3PO=now PL

‘it seems she came out of the river downstream from them’

[ML Basket Ogress, line 205]

and, by extension, to introduce the point of origin with verbs expressing extraction:

(178) a. gʷəl lild tul’ʔal tiʔəʔ skəkiʔ tiʔəʔ diʔəʔ č’ač’as

gʷəl lil–d tul’–ʔal tiʔəʔ skəkiʔ tiʔəʔ diʔəʔ č’ač’as

SCONJ be.far–ICS CNTRFG–at PROX cradleboard PROX here child

‘they remove the boy from the cradle board’

[HM Star Child, line 39]

b. huy, x%əcədaxʷ əlgʷəʔ tul’ʔal tiʔəʔ cədiɬ qʷiqʷqʷistay’bixʷ

huy x%əc–əd=axʷ əlgʷəʔ tul’–ʔal tiʔəʔ cədiɬ qʷi–qʷ–qʷistay’bixʷ

SCONJ extract–ICS=now PL CNTRFG–at PROX he ATTN–ATTN–dwarves

‘then they pulled them [the quills] out of these dwarves’

(Hess 2006: 63, line 502)

c. gʷəl ʔuxʷəcəd tul’ʔal tsiʔiɬ dəxʷəscaq’tubs

gʷəl ʔu–xʷəc–əd tul’–ʔal tsiʔiɬ dəxʷ=ʔəs–caq’–txʷ–b=s

SCONJ PFV–remove–ICS CNTRFG–at DIST:FEM ADNM=STAT–impaled–ECS–PASS=3PO

‘he took her off from where she had been impaled’

(Hess 2006: 20, line 200)

tul’ʔal is also used in a static sense to express point of origin:

(179) tul’ʔal čəd dᶻidᶻəlal’ič

tul’–ʔal čəd dᶻidᶻəlal’ič

CNTRFG–at 1SG.SUB Seattle

‘I’m from Seattle’

(Bates, Hess & Hilbert 1994: 6)

And, by extension, it is used to express how things are made or from what they are derived:

(180) a. x%ʷul’əxʷ ʔukʷəd(d)xʷ tiʔəʔ diʔəʔ bəčəɬs bədaʔ tul’ʔal tiʔəʔ tuscəcikʷ ʔup’ic’id

x%ʷul’=əxʷ ʔu–kʷəd–dxʷ tiʔəʔ diʔəʔ bə=s=čəɬ=s bədaʔ

only=now PFV–take–DC PROX here ADD=NM=make=3PO child

tul’–ʔal tiʔəʔ tu=scəcikʷ ʔu–p’ic’i–d

CNTRFG–at PROX PAST=diaper PFV–wrung–ICS

‘she took this child made from the wrung out diaper’

[HM Star Child 69, line 69]

184

b. ʔabyitəb ʔə kʷədiʔ stab suc’uqʷ[u]təbs tul’ʔal tiʔəʔ cədiɬ stab sʔuladxʷ

ʔab–yi–t–əb ʔə kʷədiʔ stab s=ʔu–c’uqʷu–t–əb=s

extend–DAT–ICS–PASS Pr REM.DMA what NM=PFV–sucked–ICS–PASS=3PO

tul’–ʔal tiʔəʔ cədiɬ stab sʔuladxʷ

CNTRFG–at PROX he what salmon

‘he was given something from a salmon to suck on’

(Hess 2006: 40, line 459)

All of these uses are quite clearly related to the basic centrifugal meaning of the preposition.

In its temporal use, tul’ʔal is used to indicate the beginning of an event or initiation of a state

or process:

(181) a. tul’ʔal kʷi tushuy ʔə tiʔəʔ swatixʷəd sʔas ʔəsɬaɬlil

tul’–ʔal kʷi tu=s=huy ʔə tiʔəʔ swatixʷəd s=ʔa=s

CNTRFG–at REM PAST=NM=be.done PR PROX land NM=be.there=3PO

ʔəs–ɬaɬlil

STAT–live

‘from the making of this world she has been living there’

(Bierwert 1996: 185, line 60)

b. diɬ suhuˑys əlgʷəʔ tul’ʔal kʷi tusʔatəbəd ʔə tiʔiɬ sqas əlgʷəʔ

diɬ s=ʔu–huy=s əlgʷəʔ tul’–ʔal kʷi tu=s=ʔatəbəd ʔə tiʔiɬ

FOC NM=PFV=be.done=3PO PL CNTRFG–at REM PAST=NM=die PR DIST

sqa–s əlgʷəʔ

older.sibling–3PO PL

‘that is what they have done since their older brother died’

(Bierwert 1996: 201, line 267)

Once again, the notion of ‘since’ is quite transparently related to the spatial notion of motion

away from.

The idea of centrifugal motion can also be seen lying behind some of the more abstract uses

of tul’ʔal, such as those in (182):

(182) a. ƛ’iw’əxʷ tul’ʔal tsiʔəʔ cədiɬ

ƛ’iw’=əxʷ tul’–ʔal tsiʔəʔ cədiɬ

escape=now CNTRFG–at PROX:FEM he

‘he escapes from her’

[ML Basket Ogress, line 49]

185

b. tux%ʷ huy ʔəsčal kʷi gʷədəxʷkʷədxʷs tul’ʔal tiʔəʔ bədaʔs

tux%ʷ huy ʔəs–čal kʷi gʷə=dəxʷ=kʷəd–dxʷ=s tul’–ʔal tiʔəʔ bədaʔ–s

just SCONJ STAT–how REM SBJ=ADNM=take–DC=3PO CNTRFG–at PROX child–3PO

‘but how could he [Coyote] manage to get her [his son’s wife] from his son?’

(Hess 2006: 23, line 17)

The use of the preposition in (182a) follows quite naturally from the notion of fleeing or

escaping as motion away from a particular place or person, while in (182b) tul’ʔal is used to

express the metaphorical movement of the son’s wife away from her husband into Coyote’s

possession.

In another case, tul’ʔal surfaces in an expression of motive:

(183) diɬəxʷ dəxʷəsaydəgʷidəxʷ ʔə tiʔəʔ ʔaciɬtabixʷ stab kʷi sdaʔs ʔə tiʔəʔ like tiʔəʔ dxʷləbiʔ,

stab kʷi sdaʔs tudiʔ sbadbadil tul’ʔal kʷi sdᶻixʷ tushuyutid ʔə sp’ic’ikʷ ʔi sɬukʷalb

diɬ=əxʷ dəxʷ=ʔəs–hay•dəgʷ–i–d=əxʷ ʔə tiʔəʔ ʔaciɬtabixʷ stab

FOC=now ADNM=STAT–know•inside–SS–ICS=now PR PROX people what

kʷi s=daʔ=s ʔə tiʔəʔ like tiʔəʔ dxʷləbiʔ stab kʷi

REM NM=name=3PO PR PROX like PROX Lummi what REM

s=daʔ=s tudiʔ sbad–badil tul’–ʔal kʷi s=dᶻixʷ

NM=name=3PO DIST.DMA DSTR–mountain CNTRFG–at REM NM=first

tu=s=huyu–t–id ʔə sp’ic’ikʷ ʔi sɬukʷalb

PAST=NM=be.done–ICD–PASS:SBRD PR Diaper.Child and moon

‘that is how they knew the names of the people like the Lummi and all the names of the

mountains there because of what Diaper Child and Moon did’

[DS Star Child, line 371]

(183) describes a situation in which the people’s state of knowledge (about the names of the

Lummi and the mountains) is said to originate from the actions of Diaper Child and Moon,

treating a cause as a metaphorical point of origin.

tul’ʔal also surfaces in some expressions of comparison:

(184) ʔiɬhaʔɬ tiʔiɬ spčuʔ tul’ʔal tiʔəʔ spčuʔ

ʔiɬ–haʔɬ tiʔiɬ spčuʔ tul’–ʔal tiʔəʔ spčuʔ

PRTV–good DIST basket CNTRFG–at PROX basket

‘that cedar-root basket is better than this cedar-root basket’

(Hess 1998: 37)

Comparative constructions are discussed further in Section 8.8.

186

2.3.1.4 Prolative liɬʔal ‘via, through’

The preposition liɬʔal is formed from the combination of ʔal with the prolative directional

particle liɬ (Section 2.7.2). Of the four spatial prepositions, liɬʔal is the least frequent and alone

has no attested temporal uses. Its basic meaning is to indicate motion along a path or location

throughout the region designated by its complement, as illustrated by the examples in (185):

(185) a. … ləʔibəš liɬʔaləxʷ tiʔəʔ tə ʔa ƛ’usəsʔux%ʷs

lə=ʔibəš liɬʔal=əxʷ tiʔəʔ tə ʔa ƛ’u=s=ʔəs–ʔux%ʷ=s

PROG=travel PRLV–at=now PROX NSPEC be.there HAB=NM=STAT–go=3PO

‘… he is traveling by the route he usually goes [along]’

(Hess 2006: 66, line 580)

b. ʔa(h) kʷi ƛ’ubəsʔiɬt’aq’ts, liɬʔal tiʔiɬ ʔiɬt’aq’t

ʔah kʷi ƛ’u=bə=s=ʔiɬ–t’aq’t=s liɬ–ʔal tiʔiɬ ʔiɬ–t’aq’t

exist REM HAB=ADD=NM=PRTV–landward=3PO PRLV–at DIST PRTV–landward

‘there he would be again up the bank, [moving] along the bank’

(Hess 1998: 65, line 16)

c. dᶻubdᶻubalikʷəxʷ əlgʷəʔ ʔal tiʔəʔ diʔəʔ, liɬʔal tiʔəʔ diʔəʔ hud

dᶻub–dᶻub–alikʷ=əxʷ əlgʷəʔ ʔal tiʔəʔ diʔəʔ liɬʔal tiʔəʔ diʔəʔ hud

DSTR–dance–ACT=now PL at PROX here PR PROX he fire

‘they danced in that place around the fire’

[AW Basket Ogress, line 48]

The adjunct adverbial phrase in each of these examples expresses an area within which the event

expressed by the verb takes place. Rather than expressing an exact location as ʔal does, liɬʔal

gives the sense of the action being distributed over or throughout a region of space. This

meaning naturally extends itself to situations that can be glossed with the English preposition

through:

(186) x%ʷuˑl’ ƛ’uləʔux%ʷ liɬʔal tiʔəʔ [s]əsluʔ tiʔəʔ ƛ’u[s]pusutəbs

x%ʷul’ ƛ’u=lə=ʔux%ʷ liɬ–ʔal tiʔəʔ səsluʔ tiʔəʔ ƛ’u=s=pusu–t–əb=s

only HAB=PROG=go PRLV–at PROX hole PROX HAB=NM=thrown–ICS–PASS=3PO

‘what was being thrown [at him] would just go through the hole [in his paddle]’

[AJ Basket Ogress, line 71]

In these cases, the action described by the verb does not take place at a static location specified

by the complement of the preposition (the hole in paddle), but rather occurs in the region around

187

that location. Note that the motion of the object in (186) excludes the use of both the centripetal

dxʷʔal and the centrifugal tul’ʔal , as it includes movement both towards and away from the

deictic centre.

The basic notion of motion through a region also surfaces in expressions of conveyance,

frequently glossed as by or by means of:

(187) a. ləq’il čəd liɬʔal ti lilud

lə=q’il čəd liɬ–ʔal ti lilud

PROG=aboard 1SG.SUB PRLV–at SPEC train

‘I’m coming by train’

b. ləst’agʷt čəd liɬʔal tə stiqiw

ləs–t’agʷt čəd liɬ–ʔal tə stiqiw

PROG.STAT–be.on.top 1SG.SUB PRLV–at NSPEC horse

‘I’m riding a horse’

(Bates, Hess & Hilbert 1994: 6)

In examples like these, the main predicate of the sentence is static (q’il ‘be aboard’ and t’agʷt ‘be

on top’) but the clause itself derives a dynamic reading of motion from the adverbial adjunct

headed by liɬʔal, which indicates the event is located throughout a region rather than at a static

point. The fact that the complement of liɬʔal is not spatially extended (a train or horse) implies

that it must be moving and, therefore, that it can be interpreted as some sort of conveyance for

the actor that is on board or on top of it.

2.3.2 General preposition ʔə

Unlike the other prepositions, the general preposition ʔə seems not to have much in the way

of lexical meaning but instead functions — like the English of — primarily as a grammatical

marker indicating a range of oblique syntactic relations. One of the most frequent uses of ʔə is to

mark NP possessors (Section 7.2), and possessor-subjects of nominalized verbs (7.4.2 and 8.1.1),

as in (188):

188

(188) a. huy, k’awdxʷəxʷ tiʔiɬ sc’aliʔ ʔə tiʔiɬ čxʷəluʔ

huy k’aw–dxʷ=əxʷ tiʔiɬ sc’aliʔ ʔə tiʔiɬ čxʷəluʔ

SCONJ touch–DC=now PROX heart PR DIST whale

‘and then he bumped into the whale’s heart’

(Hess 1995: 140, line 16)

b. diɬ day’ ƛ’uscut ʔə tiʔiɬ sčətxʷəd

diɬ day’ ƛ’u=s=cut ʔə tiʔiɬ sčətxʷəd

FOC only HAB=NM=say PR DIST bear

‘that was Black Bear’s habitual saying’

(Hess 1995: 145, line 47)

It is also frequently found introducing agentive complements (8.1.7),

(189) huy x%ədtəbaxʷ ʔə tiʔəʔ cədiɬ sʔubʔubədiʔ

huy x%əd–t–əb=axʷ ʔə tiʔəʔ cədiɬ sʔub–ʔubədiʔ

SCONJ pushed–ICS–PASS=now PR PROX he DSTR–hunter

‘so then he was shoved aside by these hunters’

(Hess 1998: 86, line 217)

and oblique arguments (Section 8.1.6):

(190) a. ʔuxʷiʔxʷiʔ haw’ə ʔə tiʔiɬ sčutxʷ

ʔu–xʷiʔxʷiʔ haw’ə ʔə tiʔiɬ sčutxʷ

PFV–forage PTCL PR DIST halibut

‘he fished for halibut’

(Hess 2006: 56, line 332)

b. gʷəč’əb ʔə kʷi ʔaciɬtabixʷ

gʷəč’–əb ʔə kʷi ʔaciɬtabixʷ

search–MD PR REM people

‘they were looking for people’

[DS Star Child, line 140]

c. x%ʷul’ab ʔə ʔəca

x%ʷul’ab ʔə ʔəca

be.like Pr I

‘he is just like me’

(Bates, Hess & Hilbert 1994: 273)

In (190), the oblique objects introduced by the preposition are part of the normal government

pattern of the verb. In other cases, ʔə is used to introduce arguments, such as instruments, which

are not normally part of the verb’s government pattern:

189

(191) a. ɬut’ilibəxʷ čələp ʔə kʷi ɬu[s]əcuucləp

ɬu=t’ilib=əxʷ čələp ʔə kʷi ɬu=s=lə=cut–c=ləp

irr=sing=now 2PL.SUB PR REM IRR=NM=PROG=say–ALTV=2PL.PO

‘you will sing with [the words] you will say to him’

(Hess 1998: 99, line 228)

b. huy č’axʷadəxʷ əlgʷəʔ ʔə tiʔiɬ x%ʷubt

huy č’axʷa–d=əxʷ əlgʷəʔ ʔə tiʔiɬ x%ʷubt

SCONJ club–ICS=now PL PR DIST paddle

‘then he clubbed them with that paddle’

(Hess 2006: 62, line 486)

Such uses are not restricted to instruments, but seem to include the introduction of any kind of

argument that is potentially associated with or implied by the verb:

(192) a. gʷəl tuʔay’waʔscutəxʷ əlgʷəʔ ʔə tiʔəʔ suq’ʷaʔs

gʷəl tu=ʔay’•waʔs–t–sut=əxʷ əlgʷəʔ ʔə tiʔəʔ suq’ʷaʔ–s

SCONJ PAST=change•pair–ICS–REFL=now PL PR PROX younger.brother–3PO

‘and he changed places with his younger brother’

[HM Star Child, line 168]

b. gʷəl t(u)asdᶻidᶻihiʔəxʷ ʔə tiʔəʔ bədaʔsəxʷ, sɬukʷalb

gʷəl tu=ʔas–dᶻidᶻihiʔ=əxʷ ʔə tiʔəʔ bədaʔ–s=əxʷ sɬukʷalb

SCONJ PAST=STAT–pregnant=now PR PROX offspring–3PO=now moon

‘and she was pregnant with her son, the moon’

[DS Star Child, line 69]

c. diɬəxʷ ƛ’udsut’ilib ʔə tiʔəʔ ƛ’udsušƛ’ap

diɬ=əxʷ ƛ’u=d=s=ʔu–t’ilib ʔə tiʔəʔ ƛ’u=d=s=ʔu–šƛ’•ap

FOC=now HAB=1SG.PO=NM=PFV–sing PR PROX HAB=1SG.PO=NM=PFV–scorch•bottom

that is my song about my bottom-scorching’

[DS Star Child, line 204]

d. huy čələp ʔuhəliʔdubuɬ ʔə tiʔəʔ shuy čəɬ

huy čələp ʔu–həliʔ–dxʷ–buɬ ʔə tiʔəʔ s=huy čəɬ

SCONJ 2PL.SUB PFV–alive–DC–1PL.OBJ PR PROX NM=be.done 1PL.PO

‘for you are the ones who healed us from what was done to us’

(Hess 2006: 66, line 595)

The same is true of the oblique possessor-argument associated with lexical suffixes (Section 4.1),

which are most regularly introduced by ʔə:

190

(193) a. putəxʷ ləq’ʷup’q’ʷup’ačiʔ ʔə tiʔəʔ sčətxʷəd

put–əxʷ lə=q’ʷup’–q’ʷup’•ačiʔ ʔə tiʔəʔ sčətxʷəd

really–now PROG=DSTR–shrivel•hand Pr PROX bear

‘Black Bear’s hands really shriveled up [from the heat]’

(Hess 1995: 154, line 81)

b. ləsaxʷəbabac ʔə tiʔəʔ qʷɬaʔy dxʷʔal tiʔiɬ sɬčis tiʔəʔ ʔaʔyəds qaw’qs

lə=saxʷəb•abac ʔə tiʔəʔ qʷɬaʔy dxʷ–ʔal tiʔiɬ s=ɬčil–s=s

PROG=jump•body PR PROX log toward–Pr DIST NM=arrive–ALTV=3PO

tiʔəʔ ʔaʔyəd–s qaw’qs

PROX companion–3PO raven

‘he is jumping over the logs until he arrives at his friend, Raven’

(Hilbert & Hess 1977: 25–26)

c. ƛ’utu(u)ʔux%ʷəxʷ ƛ’utušƛ’ap ʔə kʷi sčəbid

ƛ’u=tu=ʔu–ʔux%ʷ=əxʷ ƛ’u=tu=šƛ’•ap ʔə kʷi sčəbid

HAB=PAST=PFV–go=now HAB=PAST=scorch•bottom PR REM fir.bark

‘she would go and she would scorch fir bark’

[DS Star Child, line 173]

Thus, ʔə seems to be used as the marker of a variety of oblique argument relations.

In addition to introducing oblique arguments, ʔə is used with certain types of adjunct

adverbial phrases (8.2.7), although in these roles it shows considerable overlap with the spatial

prepositions, especially ʔal. Thus, ʔə is found not infrequently in expressions of location where

we might have expected ʔal or dxʷʔal:

(194) a. bələsq’iltubəxʷ ʔə tiʔəʔ cədiɬ q’il’bids əlgʷəʔ

bə=ləs–q’il–txʷ–b=əxʷ ʔə tiʔəʔ cədiɬ q’il’bid–s əlgʷəʔ

ADD=PROG.STAT–board–ECS–PASS=now PR PROX he canoe–3PO PL

‘they got [things] loaded again into their canoe’

(Hess 2006: 70, line 681)

b. tiʔəʔ sp’ic’ikʷ gʷəl ƛ’ax%ʷ x%ʷul’ab ʔə tiʔəʔ ƛ’usušəqcut ʔə tiʔəʔ sɬukʷaɬ

tiʔəʔ sp’ic’ikʷ gʷəl ƛ’ax%ʷ x%ʷul’ab ʔə tiʔəʔ

PROX Diaper.Child SCONJ grow in.the.same.way PR PROX

ƛ’u=s=ʔu–šəq–t–sut ʔə tiʔəʔ s–ɬukʷaɬ

HAB=NM=PFV–be.high–ICS–REFL PR PROX NP–sun

‘Diaper Child, he grows up and it seems that he lifts himself up to the Sun’

[DS Star Child, line 161]

The word qaw’qs is given in the original source as kaw’qs; Hess (p.c.) now considers this an error.

191

c. gʷəl tux%ʷəxʷ əlgʷəʔ x%ʷul’əxʷ ʔukʷaʔtəb ʔə tiʔəʔ q’il’bids əlgʷəʔ

gʷəl tu=ʔux%ʷ=əxʷ əlgʷəʔ x%ʷul’=əxʷ ʔu–kʷaʔ–t–b ʔə tiʔəʔ

SCONJ PAST=go=now PL only=now PFV–release–ICS–PASS PR PROX

q’il’bid–s əlgʷəʔ

canoe–3PO PL

‘and, therefore, he simply let go of them in [i.e., along with] their canoe’

(Hess 2006: 70, line 663)

It is also found in expression of path or distributed location where ʔal or liɬʔal are also found:

(195) a. hiqab čəd ʔuʔibəš ʔə tə lil

hiqab čəd ʔu–ʔibəš ʔə tə lil

excessively PR PFV–walk PR SPEC far

‘I walked too far’

(Hess & Hilbert 1976: I, 51)

b. gʷəl dxʷtəyiləxʷ ʔə tiʔiɬ stuləkʷ

gʷəl dxʷ–təyil=əxʷ ʔə tiʔiɬ stuləkʷ

SCONJ CTD–go.upriver=now PR DIST river

‘and he travels up the river’

(Hess 1995: 153, line 49)

c. ƛ’ub čəɬ ɬugʷax%ʷ ʔə kʷi ɬudəxʷgʷax%ʷ čəɬ

ƛ’ub čəɬ ɬu=gʷax%ʷ ʔə kʷi ɬu=dəxʷ=gʷax%ʷ čəɬ

okay 1PL.SUB IRR=wander PR REM IRR=ADNM=wander 1PL.PO

‘we shall walk where we shall walk’

(Hess 1998: 101, line 266)

Temporal uses of ʔə are also attested, as, for example, in certain expressions of duration or

temporal setting:

(196) a. ɬuhuyil sq’axʷ dxʷʔal gʷəsxʷiʔs gʷə(s)šəd¸aldubuts ʔə kʷi haʔkʷ

ɬu=huy–il s–q’axʷ dxʷ–ʔal gʷə=s=xʷiʔ=s

IRR=be.done–INCH NP–frozen CNTRPT–at SBJ=NM=NEG=3PO

gʷə=s=šədᶻal–dxʷ–but=s ʔə kʷi haʔkʷ

SBJ=NM=go.outdoors–DC–REFL=3PO PR REM long.time

‘it will become ice so that he cannot get himself out of the house for a long time’

(Hess 1998: 101, line 269)

b. gʷəl dəgʷi kʷi ɬuʔibəš ʔə kʷi ɬax%

gʷəl dəgʷi kʷi ɬu=ʔibəš ʔə kʷi ɬax%

SCONJ you REM IRR=travel PR REM dark

‘and you will be the one who will travel at night [i.e. while it is dark]’

[MW Star Child, line 107]

192

c. ƛ’ub dəgʷi kʷi ɬuləʔibəš ʔə kʷi dxʷcqʷuɬ

ƛ’ub dəgʷi kʷi ɬu=lə=ʔibəš ʔə kʷi dxʷcqʷuɬ

well you REM IRR=PROG=travel PR REM daytime

‘you should be the one to travel during the day’

[MW Star Child, line 115]

ʔə is also found introducing phrases describing events that are in progress at the time the event

expressed by the main clauses occurs:

(197) a. x%ʷul’ ləcuʔiʔɬadəb ʔə tiʔəʔ sq’ʷəlaɬəd ʔə tiʔiɬ ƛ’usq’ʷəls

x%ʷul’ ləcu–ʔi–ʔɬad–əb ʔə tiʔəʔ sq’ʷəlaɬəd ʔə tiʔiɬ ƛ’u=s=q’ʷəl=s

only CONT–ATTN–eat–PASS PR PROX berry PR PROX HAB=NM=ripe=3PO

‘he was [simply] eating berries [directly off the bush] as they ripened’

(Hess 1995: 143, line 20)

b. ʔəšuɬəxʷ əlgʷəʔ ʔə tiʔiɬ sɬčisəbəxʷ

ʔəs–šuɬ=əxʷ əlgʷəʔ ʔə tiʔiɬ s=ɬčil–s–əb=s=əxʷ

STAT–see=now PL PR DIST NM=arrive–ALTV–PASS=3PO=now

‘they

are watching as they

are come after’

(Hess 2006: 61, line 466)

It is also found with adverbial adjunct phrases expressing a point in time when an event occurs:

(198) a. hay, diɬ dəxʷhuyuds ʔušəbabdxʷ tsiʔəʔ ʔalš ʔə tiʔəʔ suʔəɬadəps

hay diɬ dəxʷ=huyu–d=s ʔušəbabdxʷ tsiʔəʔ ʔalš

SCONJ FOC ADNM=be.done–ICS=3PO poor.dear PROX:FEM cross.sex.sibling

ʔə tiʔəʔ s=ʔu–ʔəɬadəp=s

PR PROX NM=PFV–give.feast=3PO

‘so, that is how he made his sister poor when she [tried to] give a feast’

(Hess 1998: 63, line 66)

b. gʷəl huy ʔux%aab ʔə kʷi ʔəscuuc

gʷəl huy ʔu–x%aab ʔə kʷi ʔəs–cuu–c

SCONJ SCONJ PFV–cry PR REM STAT–say–ALTV

‘and then he cried when they told him’

[JS Basket Ogress, line 21]

This third-person possessive marker is missing from the last word this example, but would be expected in more

measured speech.

193

c. ʔux%ʷcəbəxʷ ʔə sp’ic’ikʷ tiʔiɬ sqas ʔə tiʔiɬ sxʷiʔs kʷi gʷat ləšəqəd

ʔux%ʷ–c–əb=əxʷ ʔə sp’ic’ikʷ tiʔiɬ sqa–s ʔə tiʔiɬ

go–ALTV–PASS=now PR Diaper.Child DIST older.brother–3PO Pr DIST

s=xʷiʔ=s kʷi gʷat lə=šəq–əd

NM=NEG=3PO REM who PROG=be.high–ICS

‘Diaper Child went to his older brother when there was no one to put him up high’

[DS Star Child, line 268]

There is also at least one example in the corpus of ʔə introducing an expression of temporal

sequence in a context where it seems to overlap with the temporal uses of ʔal:

(199) a. tiləbəxʷ kʷi scuts ʔə kʷi hagʷagʷəxʷ

tiləb=əxʷ kʷi s=cut=s ʔə kʷi hagʷ–agʷ=əxʷ

immediately=now REM NM=say=3PO PR REM DIM.EFF–long.time=now

‘after a long time he said’

[JS Basket Ogress, line 42]

b. hay ʔəɬəd tiʔiɬ sčətxʷəd ʔal suʔəɬəds tiʔiɬ sʔuladxʷ

hay ʔəɬəd tiʔiɬ sčətxʷəd ʔal s=ʔu–ʔəɬəd=s tiʔiɬ sʔuladxʷ

SCONJ eat DIST bear at NM=PFV–eat=3PO DIST salmon

‘so Black Bear ate after he had eaten the salmon’

(Hess 1995: 152, lines 22 – 23)

As with the overlap in spatial domain, it is unclear from the context of the attested sentences and

their translations what, if any, semantic contrast there is between the two prepositions in these

situations. ʔə appears to be the less specific of the two. The literal meaning of ʔal has an obvious

semantic relationship to its uses in these circumstances, whereas ʔə has a more generic flavour,

perhaps only signaling the adjunct role of the adverbial PP and leaving the exactly semantic link

between the two elements to be inferred by the hearer from context.

Just as ʔə is found infringing on the spatial and temporal uses of the other prepositions, it is

also found overlapping with their more abstract uses, such as expressions of manner:

(200) a. ƛ’ubəxʷ čələp ʔəšuuc tiʔiɬ ʔacʔaciɬtalbixʷ ʔə kʷi haʔɬ

ƛ’ub=əxʷ čələp ʔəs–šuu–c tiʔiɬ ʔac–ʔaciɬtalbixʷ ʔə kʷi haʔɬ

okay=now 2PL.SUB STAT–see–ALTV PROX DSTR–person PR REM good

‘you folks should look after those people well’

(Hess 2006: 65, line 563)

194

b. gʷəl lələč’, čəxʷə x%əqəd ʔal dəgʷi ʔə kʷi ƛ’əq’ʷ

gʷəl lə=ləč’ čəxʷə x%əq–əd ʔal dəgʷi ʔə kʷi ƛ’əq’ʷ

SCONJ ATTN–fill 2SG.COORD wrap–ICS PR you PR REM watertight

‘fill it up a bit and you wrap it to yourself so that it is watertight’

[JS Basket Ogress, line 48]

c. xʷiʔ kʷ adsaʔliʔɬ ʔə kʷi gʷəɬ stubš

xʷiʔ kʷi ad=s=haʔl•iʔɬ ʔə kʷi gʷəɬ stubš

NEG REM 2SG.PO=NM=stop.crying•child PR REM ASSC man

‘you will not comfort the child as a man (i.e., in the manner appropriate to males)’

[DS Star Child, line 96]

d. … čəda ɬutx%ʷud tiʔəʔ x%ʷiləb ʔə kʷi buusaɬ čxʷa ɬukʷaʔəd

čəda ɬu=tx%ʷu–d tiʔəʔ x%ʷiləb ʔə kʷi buus•aɬ

1SG.COORD IRR=pull–ICS PROX rope PR REM four•times

čxʷa ɬu=kʷaʔ–əd

2SG.COORD IRR=released–ICS

‘and I will pull in the rope four times and you will release it’

[DS Star Child, line 54]

It is also found introducing expressions of motive (as opposed to expressions of purpose, which

are more typically introduced by dxʷʔal),

(201) a. haˑy bədukʷtubəxʷ ʔə tiʔiʔiɬ ʔə tiʔəʔ suhuys

hay bə=dukʷ–tu–b=əxʷ ʔə tiʔ–iʔiɬ ʔə tiʔəʔ s=ʔu–huy=s

INTJ ADD=anger–ECS–PASS=now PR PL–DIST PR PROX NM=PFV–be.done=3PO

‘so they got mad at him for what he had done’

(Hess 1998: 89, line 298)

b. kʷədad tiʔəʔ qʷuʔ [tiʔiɬ] dəxʷuc’agʷačiʔbs ʔə tiʔəʔ sbaɬs

kʷəda–d tiʔəʔ qʷuʔ tiʔiɬ dəxʷ=ʔu–c’agʷ•ačiʔ–b=s

take–ICS PROX water DIST ADNM=PFV–wash•hand–MD=3PO

ʔə tiʔəʔ s=baɬ=s

PR PROX NM=cure=3PO

‘he took [some] water to wash his hands for the curing-ceremony’

(Hess 1998: 57, line 36)

and cause:

195

(202) a. tiləb ləʔatəbəd tiʔəʔ qʷiqʷqʷistay’bixʷ ʔə tiʔiɬ sc’uqʷəb ʔə tiʔiɬ st’uʔq’ʷ ʔə tiʔiɬ buʔqʷ

tiləb lə=ʔatəbəd tiʔəʔ qʷi–qʷ–qʷistay’bixʷ ʔə tiʔiɬ sc’uqʷəb

suddenly PROG=die PROX ATTN–ATTN–dwarves PR DIST quill

ʔə tiʔiɬ st’uʔq’ʷ ʔə tiʔiɬ buʔqʷ

PR DIST feather PR DIST waterfowl

‘right away the dwarves died from the quills of those feathers of those Duck People’

(Hess 2006: 62, line 476)

b. gʷəl hikʷ ʔuhiiɬ əlgʷəʔ ʔə tiʔəʔ shəliʔdubs əlgʷəʔ ʔə tiʔiɬ sxʷəctəbs

gʷəl hikʷ ʔu–hiiɬ əlgʷəʔ ʔə tiʔəʔ s=həliʔ–dxʷ–b=s əlgʷəʔ

SCONJ big PFV–happy PL PR PROX NM=alive–DC–PASS=3PO PL

ʔə tiʔiɬ s=xʷəc–t–əb=s

PR DIST NM=extract–ICS–PASS=3PO

‘and they were very glad for having been saved by the removal [of the quills]’

(Hess 2006: 65, line 560)

In this last function, ʔə serves a role that closely parallels its use introducing agentive

complements, which are in a very direct sense often the causes of events expressed by passivized

verbs.

Although phrases headed by ʔə are nearly identical (and often interchangeable) with phrases

headed by other prepositions, the preposition itself does not share the other syntactic properties

of prepositions discussed above, such as the ability to host clitics, nor can a prepositional phrase

headed by ʔə itself serve as the head of a predication. One property that it does share with the

other prepositions is the ability to take an NP complement that lacks a determiner, but these

cases are limited to a few very specific circumstances, such as complements that are personal

pronouns or proper names:

(203) a. ʔux%ʷc čəxʷ kʷi č’ƛ’aʔ x%ʷulab ʔə dəgʷi kʷi sx%əbs čxʷa ʔəƛ’txʷ

ʔux%ʷ–c čəxʷ kʷi č’ƛ’aʔ x%ʷulab ʔə dəgʷi kʷi s=x%əb=s

go–ALTV 2SG.SUB REM stone thusly PR you REM NM=heavy=3PO

čxʷa ʔəƛ’–txʷ

2SG.COORD go–ECS

‘go get a stone that is like you in weight and bring it’

[AW Basket Ogress, line 79]

196

b. tuyəyəhubid tsi tusk’ʷuy ʔə Martin

tu=yəyəhub–bi–d tsi tu=sk’ʷuy ʔə Martin

PAST=tell.story–MAP–ICS SPEC:FEM PAST=mother PR Martin

‘it was told to him

by Martin

’s late mother’

[DS Star Child, line 8]

It should be noted that sentences of all these types are also attested with determiners in the same

environments. The same is true of ʔə-phrases whose complements are place names or the names

of personified characters in myths such as Sun and Moon, which can also optionally lack a

determiner:

(204) a. … dəxʷʔahəxʷ ʔə tiʔəʔ town ʔə La Conner

dəxʷ=ʔa=həxʷ ʔə tiʔəʔ town ʔə La Conner

ADNM=be.there=now PR PROX town PR La Conner

‘… why there is a town of La Conner’

[MS Basket Ogress, line 80]

b. sk’ʷuy ʔə sɬukʷalb tsiʔəʔ cisxʷixʷt’il

sk’ʷuy ʔə sɬukʷalb tsiʔəʔ cisxʷixʷt’il

mother PR moon PROX:FEM Cisxwixwt’il

‘Cisxwixwt’il was the mother of Moon’

[DS Star Child, line 4]

Unlike the spatial prepositions, however, ʔə does not take complements of other types without

determiners.

2.4 Determiners

Like many Salishan languages, Lushootseed has a complex determiner system which plays

an essential role in the grammar. Determiners are the obligatory initial element in almost every

referential expression, and presence or absence of a determiner is frequently the only way to

identify whether the syntactic role of a particular element in a clause is that of predicate or

argument (for further discussion, see Section *.*). Determiners also play an essential role in

reference-tracking and the marking of communicative structure in discourse (Section 11). This

section will concentrate on the meanings and general syntax of determiners, leaving more

197

detailed discussion of their specific functions to the relevant sections in other parts of the

grammar.

The full set of Lushootseed determiners is given in Table 58:

specific

demonstrative

non-demon

prox

dist

rem

specific

unique

non-specific

masc

tiʔəʔ (NL)

ti (SL)

tiʔiɬ (NL)

tiiɬ (SL)

kʷi

ti (NL)

šə (SL)

tiʔacəc

tə

fem

tsiʔəʔ (NL)

tsi (SL)

tsiʔiɬ (NL)

tsiiɬ (SL)

kʷsi

tsi

sə (SL)

tsiʔacəc

tsə

tiʔiʔəʔ

tiʔiʔiɬ

—

Table 58: Lushootseed determiners

The system is notable for the number of semantic distinctions it makes. The highest level of

these seems to be between what will be called here specific vs. non-specific. In Lushootseed,

this distinction is drawn between those determiners that single out a specific entity from amongst

the set of entities designated by the referring expression, and those that do not. The latter set has

only two members, the non-feminine and feminine non-specific determiners, tə and tsə. Within

the former set, a further distinction is drawn between the demonstrative determiners, which in

addition to singling out a specific entity also locate it in space or time, and the non-

demonstratives, which do not specify a location. The demonstrative determiners specify three

degrees of spatial deixis — proximal, distal, and remote — and the non-demonstratives

distinguish between a simple specific determiner, ti, and another form which encodes an

additional category, uniqueness. In addition to the plethora of deictic categories, this set of

determiners also encodes distinctions of natural (i.e., biological) gender and, to a lesser extent,

number.

As their name implies, the primary syntactic function of determiners is to introduce

referential expressions such as NPs and the complements of prepositions, as in (205):

This form is quite often reduced to [tiiɬ] in rapid or causal speech and is often rendered as [tiʔiɬə] in Skagit.

David Beck 10-2-7 2:19 PM

Comment: note the etymological

relationship between ti?acec and ?acec

198

(205) a. gʷəl ɬaliləxʷ ti stubš

gʷəl ɬalil=əxʷ ti stubš

SCONJ come.ashore=now SPEC man

‘then the man comes ashore’

[JS Basket Ogress, line 72]

b. ʔəsɬaɬlil kʷsi bšč’ad ʔal tə hikʷ ʔalʔal

ʔəs–ɬaɬlil kʷsi bšč’ad ʔal tə hikʷ ʔalʔal

STAT–live REM:FEM louse at NSPEC big house

‘Louse lives in a big house’

[EK Lady Louse, line 1]

c. ɬusʔəɬəds tiʔiʔəʔ stawixʷaʔɬ

ɬu=sʔəɬəd–s tiʔ–iʔəʔ stawixʷaʔɬ

IRR=food–3PO PL–PROX children

‘the children will be her food’

[MS Basket Ogress, line 28]

Normally, referential expressions in Lushootseed, require a determiner (or a demonstrative

adverbial — Section 2.5.1).The presence or absence of a determiner introducing a phrase is often

the only way in which to determine whether its syntactic role is that of an actant or that of a

predicate. For example, the sentence in (205c) contains two nouns — sʔəɬəd ‘food’ and

stawixʷaʔɬ ‘children’; the fact that stawixʷaʔɬ is introduced by a determiner identifies it as the

subject, while the absence of a determiner identifies sʔəɬəd as the predicate.

These

constructions will be discussed in more detail in Section 8.3.1.

Unlike English articles, Lushootseed determiners are also used with proper names and

personal pronouns when these are subjects or objects of a verb:

(206) a. ɬuʔibəš tsiʔiɬ ɬx%ʷubx%

ɬuʔibəš tsiʔiɬ ɬx%ʷubx%

IRR=travel DIST:FEM Łx%ʷubx%

‘Łx%ʷubx% will just travel’

[ML Basket Ogress, line 180]

Louse in this case is used as a female proper name in a traditional story.

The presence of the tense-marker, ɬu=, is not a good indicator of predicate-hood, as this clitic can appear on both

predicative and non-predicative elements. See Section *.* for discussion.

199

b. ləcukʷaxʷad čəɬ ti dəgʷi

ləcukʷaxʷad čəɬ ti dəgʷi

CONT–help–ICS 1PL.SUB SPEC you

we are helping you’

[MS Basket Ogress, line 65]

With proper names, the use of the determiners seems to be obligatory; with the personal

pronouns, they are almost always used to introduce subjects and objects, although there are one

or two exceptions in the corpus. Neither proper names nor personal pronouns require a

determiner when used vocatively or as appositives:

(207) a. ɬučubə čəd, q%əlqʷəlwič

ɬu=čubə čəd qʷəlqʷəlwič

IRR=go.inland 1SG.SUB Qʷəlqʷəlwič

‘I will go inland, Qʷəlqʷəlwič’

(Hess 1998 84, line 174)

b. gʷəl cick’ʷəxʷ ʔəstaˑgʷəxʷəxʷ tsiʔəʔ cədiɬ ʔay’əds, ɬx%ʷubx%

gʷəl cick’ʷ=əxʷ ʔəs–taˑgʷəxʷ=əxʷ tsiʔəʔ cədiɬ ʔay’əd=s ɬx%ʷubx

SCONJ very=now STAT–hungry=NOW PROX:FEM she companion=3PO Łx%ʷubx%

‘and her companion, Łx%ʷubx%, was very hungry’

[ML Basket Ogress, line 148]

c. ʔəbil’ čəxʷ ɬuʔitut čɬa ɬugʷəlalcid, dəgʷi

ʔəbil’ čəxʷ ɬu=ʔitut čɬa ɬu=gʷəlal–t–sid dəgʷi

if 2SG.SUB IRR=sleep 2SG.COORD IRR=punished–ICS–2SG.OBJ you

‘you, if you fall asleep we will punish you!’

[AW Basket Ogress, line 64]

Appositive nouns almost invariably lack determiners, as discussed in Section *.* below.

Determiners not only introduce simple noun phrases, they also introduce headless relative

clauses (Section 7.4.1) and nominalized clauses (7.4.2):

(208) a. ʔuˑ, ʔa tiʔəʔ ʔuqʷiʔaacəbš

ʔu ʔa tiʔəʔ ʔu–qʷiʔaad–c–bš

INTJ be.there PROX PFV–call.out–ALTV–1SG.OBJ

‘so, there is one who called me’

[AJ Basket Ogress, line 35]

b. diɬ tiʔəʔ səxʷəlušəd tiʔəʔ sʔabyitəbs tiʔəʔ kikəwič

diɬ tiʔəʔ səxʷəlušəd tiʔəʔ s=ʔab–yi–t–əb=s

FOC PROX fish.tail PROX NM=extend–DAT–ICS–PASS=3PO

200

tiʔəʔ ki–kəwič

PROX ATTN–hunchback

‘what was given to Little Hunchback was the fish-tail’

[AJ Basket Ogress, line 29]

Most of the determiners can also be used as independent third-person pronouns:

(209) a. kʷədatəb tiʔiɬ

kʷəda–t–b tiʔiɬ

taken–ICS–PASS DIST

‘that one was taken’

(Hess 2006: 59, line 428)

b. ɬuhudyid čəɬ tiʔacəc

ɬu=hud–yi–d čəɬ tiʔacəc

PAST=burn–DAT–ICS 1PL.SUB UNQ

‘we will make a fire for these very ones’

(Hess 2006: 64, line 535)

c. x%əɬ ti ƛ’ascɬil tiʔəʔ ʔal ti

x%əɬ ti ƛ’u=ʔas–cɬil tiʔəʔ ʔal ti

seemingly HAB=STAT–bleed PROX at SPEC

‘it is just as if this one is bleeding in this [place]’

(Hess 2006: 15, line 105)

The only two determiners not attested in this use are the remote demonstrative, kʷi/kʷsi, and the

non-specifics tə/tsə. The failure of the latter to function as pronouns is not surprising, as they are

used to identify a general type of entity rather than a specific individual, and so are not a good

candidate for use as an anaphor. The absence of the former in a pronominal role may be related

to its association with hypothetical and backgrounded elements in discourse. It should also be

noted that the use of the specific determiner ti as a pronoun is relatively infrequent, although it is

attested enough in the corpus that it has to be considered a regular grammatical construction.

When used as pronouns, the determiners are amenable to modification by relative clauses,

either embedded copular constructions such as those shown in (210a) or relative clauses with

verbal predicates themselves headed by determiners, as in (210b):

The use of determiners to introduce relative clauses is discussed in Section 7.4.1.

201

(210) a. gʷəl, ʔal tiʔəʔ dbədaʔ tiʔəʔ ti adƛ’əwc’lax%ad, tiʔəʔ ti adɬidšəd, tiʔəʔ ti adstab, tiʔəʔ

adbitbitaʔs

gʷəl ʔal tiʔəʔ d–bədaʔ tiʔəʔ ti ad–ƛ’əwc’lax%ad tiʔəʔ ti

SCONJ at PROX 1SG.PO–child PROX SPEC 2SG.PO–arm.decorations PROX SPEC

ad–ɬidšəd tiʔəʔ ti ad–stab tiʔəʔ ad–bit–bitaʔs

2SG.PO–leg.tie PROX SPEC 2SG.PO–what PROX 2SG.PO–DSTR–breechcloth

‘so, here, my son, are your armbands, your leg-ties, your things, your breechcloths’

(lit. ‘so this [which is] the-your

armband, this [which is] the-your

leg-tie, this [which

is] the-your

thing, this [which is] the-your

breechcloth, are at my son

’)

(Hess 2006: 35, line 322)

b. gʷəl gʷəgʷiidəxʷ čələp ti ʔal tiʔiɬ ti luƛ’ ləʔibəš liɬʔaləxʷ tiʔəʔ tə ʔa ƛ’usəsʔux%ʷs

gʷəl gʷə=gʷii–d=əxʷ čələp ti ʔal tiʔiɬ ti luƛ’ lə=ʔibəš

SCONJ SBJ=invite–ICS=now 2PL.SUB SPEC at PROX SPEC old STAT–travel

liɬ–ʔal=əxʷ tiʔəʔ tə ʔa ƛ’u=s=ʔəs–ʔux%ʷ=s

PRLV–at=now PROX NSPEC be.there HAB=NM=STAT–go=3PO

‘and you guys could invite him

when that one [who is] the old fellow

who travels goes

by [this place] here where he is usually going’

(Hess 2006: 66, line 580)

Although such constructions seem somewhat convoluted from an English perspective, they are

considered good Lushootseed style.

In addition to introducing the arguments of nouns and the complements of prepositions,

determiners are found introducing adjunct adverbial clauses such as those in (211):

(211) a. gʷəl kʷədub tiʔiɬ x%iʔdubəs ʔə tsiʔiɬə

gʷəl kʷəd–dxʷ–b tiʔiɬ x%iʔ–dxʷ–b=əs ʔə tsiʔiɬə

SCONJ taken–DC–PASS DIST caught.red.handed–DC–PASS=3SBJ PR DIST:FEM

‘they would be grabbed if they were caught in the act by her’

[DM Basket Ogress, line 63]

b. dᶻəlulčbidəxʷ tiʔəʔ hikʷ hud tiʔəʔ dəxʷuǰuʔils

dᶻəl•ulč–bi–d=əxʷ tiʔəʔ hikʷ hud tiʔəʔ dəxʷ=ʔu–ǰuʔil=s

turn•belly–MAP–ICS=now PROX big fire PROX ADNM=PFV–rejoice=3PO

‘she turned her belly to the big fire because she was happy’

[ML Basket Ogress, line 78]

In sentences like these, the import of the determiners is that they introduce clauses describing

specific (rather than general types of) events; the determiner also serves as a marker of

subordination, indicating that the adverbial clause is a modifier of the preceding predicate rather

202

than a coordinate element. This type of expression will be discussed in more detail in Section

9.5.

The largest set of determiners is the specifc determiners. All of the members of this set serve

to single out specific entities of the type defined by the expression they introduce. In doing so,

the determiner establishes a referent for that expression which can be identified by the hearer.

Within the set of specific determiners, the demonstrative determiners are distinguished from the

non-demonstratives in that the former identify their referent by locating or grounding them in

physical or discourse space, assigning them one of three degrees of distance from the speaker.

The demonstrative determiners are tiʔəʔ ‘PROXIMAL’, tiʔiɬ ‘DISTAL’, and kʷi ‘REMOTE’,

illustrated in (212) in their literal, spatial uses:

(212) a. ɬugʷəx%aliǰəd čəɬ tiʔəʔ dsheliʔdub ʔə ti sʔubʔubədiʔ

ɬu=gʷəx%•aliǰ–əd čəɬ tiʔəʔ d=s=heliʔ–dxʷ–b ʔə

IRR=untie•bundle–ICS 1PL.SUB PROX 1SG.PO=NM=alive–DC–PASS PR

ti sʔub–ʔubədiʔ

SPEC DSTR–hunter

‘we will unpack this that I was given by some hunters’

(Hess 1998: 82, line 126)

b. x%ʷul’əxʷ čəd ƛuʔahaʔ tiʔiɬ ƛ’udsq’əlb lil’lil

x%ʷul’=əxʷ čəd ƛu=ʔahaʔ tiʔiɬ ƛ’u=d=s=q’əlb lil’–lil

only=now 1SG.SUB IRR=right.there DIST IRR=1SG.PO=NM=make.camp DSTR–be.far

‘I would just be right there in that place I was camping a little way off’

(Hess 2006: 34, line 292)

c. ɬuləʔux%ʷ čəd dxʷʔal kʷi siʔab ʔabsləx%il

ɬu=lə=ʔux%ʷ čəd dxʷʔal kʷi siʔab ʔas–bəs–ləx%il

IRR=PROG=go 1SG.SUB PR REM noble STAT–PROP–daylight

‘I will go to the nobleman who has the daylight (i.e., the sun)’

(Hilbert & Hess 1977: 19)

The proximal determiner, tiʔəʔ, is used as in (212a) to refer to something in the immediate

vicinity of the speaker, while the distal determiner, tiʔiɬ, is used to refer to something at a

distance from the speaker (212b). In its spatial use, the remote determiner, kʷi, indicates that

something is at a greater distance from the speaker, usually out of sight (212c). The spatial uses

203

of the demonstrative determiners are rather imprecise in terms of absolute distance from the

speaker, and these words — particularly tiʔəʔ and tiʔiɬ — can be used interchangeably in certain

circumstances, depending on context and the intention of the speaker.

In addition to being used to locate objects in physical space, the demonstrative determiners

are also occasionally used in a metaphorical sense to indicate distance in time, although temporal

expressions seem only to maintain a consistent two-way distinction between present-time,

immediate events (or those events that are certain to occur soon), and events in the future. In the

first case, the proximal and distal determiners, tiʔəʔ and tiʔiɬ, are used, as in (213):

(213) a. tiʔəʔ dəč’uʔ syəyəhub q’iq’x%ʷuʔ tiʔəʔ ʔuʔilid čəd ʔal tiʔəʔ səɬax%il

tiʔəʔ dəč’uʔ syəyəhub q’iq’x%ʷuʔ tiʔəʔ ʔu–ʔili–d čəd

PROX one legend short PROX PFV–sing–ECS 1SG.SUB

ʔal tiʔəʔ səɬax%il

at PROX evening

‘what I told you this evening [is] this one short story’

[HM Star Child, line 194]

b. huy ləx%iləxʷ ʔal tiʔiɬ ɬuˑp

huy ləx%–il=əxʷ ʔal tiʔiɬ ɬup

SCONJ light–INCH=now at DIST dawn

‘then, it became light at dawn’

(Hess 1998: 101, line 274)

c. tiʔiɬ ʔalʔals ƛ’udəxʷʔux%s ʔal tiʔiɬ pədt’əs

tiʔiɬ ʔalʔal–s ƛ’u=dəxʷ=ʔux%=s ʔal tiʔiɬ pədt’əs

DIST house–3PO IRR=ADNM=go=3PO at DIST winter

‘his house where he would go in the winter’

(Hess 1995: 143, line 7)

The use of the proximal determiner to refer to the present-time, immediate experience of the

speaker in (213a) is fairly transparent; however, in narrative, the distal determiner is far more

frequently used to refer to present-time events, as shown in (213b), perhaps because these events

are present-time relative to the event under discussion rather than the speech act.

Although past-time events can be introduced in stories with the distal determiner, they are

also found with the proximal determiner and, even more frequently, the remote determiner, kʷi:

204

(214) a. x%ʷul’ul’əxʷ p’q’ac ʔal kʷi tusɬčilsəxʷ

x%ʷul’–ul’=əxʷ p’q’ac ʔal kʷi tu=s=ɬčil–s=əxʷ

INTNS–only=now rotten.wood at REM PAST=NM=arrive–ALTV=now

‘it was nothing but a rotten log when he arrived’

(Hess 1998: 88, line 269)

b. tiʔəʔ swatixʷtəd gʷəl tasbəsad ʔal kʷi tuhaʔkʷ

tiʔəʔ swatixʷtəd gʷəl tu=ʔas–bəsad ʔal kʷi tu=haʔkʷ

PROX country SCONJ PAST=STAT–dark at REM PAST=long.time

‘the land was dark in the distant past’

(Hilbert & Hess 1977)

Determiner choice seems to be fairly free in temporal expressions of past-time events, at least in

narrative, and is probably driven largely by issues of reference-tracking, topicality,

backgrounding/foregrounding, and the relative salience of the events in their discourse context.

More consistently, the remote determiner kʷi is found with expressions of future events,

particularly those whose realization is uncertain or whose precise time of realization is unknown:

(215) a. day’ čələp ɬuʔabaqtəb ʔal kʷi pədtab ʔal kʷi gʷəgʷiid čəɬ ti čxʷəluʔ

day’ čələp ɬu=ʔabaq–t–əb ʔal kʷi pədtab ʔal kʷi

only 2PL.SUB IRR=return–ICS-PASS at REM when at REM

gʷə=s=gʷii–d čəɬ ti čxʷəluʔ

SBJ=NM=invite–ICS 1PL.SUB SPEC whale

‘indeed, you folks will be returned when we invite Whale’

(Hess 2006: 78, line 878)

b. x%ʷul’əxʷ čəxʷ ɬuʔah kʷi ɬadsucucut ʔal kʷi ɬuč’itəxʷ ɬuʔaciɬtalbixʷ

x%ʷul’=əxʷ čəxʷ ɬu=ʔah kʷi ɬu=ad=s=ʔu–cut–cut ʔal kʷi

only=now 2SG.SUB IRR=be.there REM IRR=2SG.PO=NM=PFV–DSTR–say at REM

ɬu=č’it=əxʷ ɬu=ʔaciɬtalbixʷ

IRR=be.near=now IRR=person

‘you will just be there talking away when the people are nearby’

(Hess 2006: 7, line 117)

c. ɬasʔaciɬ čəd dxʷʔal kʷi ɬadsʔəƛ’txʷ

ɬu=ʔas–ʔaciɬ čəd dxʷ–ʔal kʷi ɬu=ad=s=ʔəƛ’–txʷ

IRR=STAT–wait 1SG.SUB CNTRPT–at REM IRR=2SG.PO=NM=come–ECS

‘I will be waiting for you to bring it’

(Hilbert & Hess 1977: 19)

205

The use of the remote determiner for expressions of the future is also clearly related to another

very frequent use of kʷi — the introduction of abstract, hypothetical, or unreal entities into

discourse, as in (216):

(216) a. ƛ’ubəxʷ čəd ɬup’ayəqəxʷ ʔə kʷi sup’qs

ƛ’ub=əxʷ čəd ɬu=p’ayəq=əxʷ ʔə kʷi sup’qs

okay=now 1SG.SUB IRR=carve.canoe=now PR REM hair.seal

‘I had better carve a hair seal’

(Hess 2006: 46, line 106)

b. gʷəl ɬuʔəbsxʷiʔxʷiʔ čəd ʔə kʷi hiˑkʷ tatačulbixʷ

gʷəl ɬu=ʔəs–bəs–sxʷiʔxʷiʔ čəd ʔə kʷi hikʷ tatačulbixʷ

SCONJ IRR=STAT–PROP–game 1SG.SUB PR REM big big.game.animal

‘and I will have as my game a very large animal’

[MW Star Child, line 74]

In (216a), the remote determiner is used to introduce the expression of an entity, the carving of a

hair seal, that does not yet exist but which is the intention of the speaker to create. In (216b), kʷi

is used to refer to an animal that has not yet been found, and which may possibly not exist.

Along the same lines, kʷi is also found in desiderative expressions:

(217) a. x%aƛ’txʷ kʷi sčəlus

x%aƛ’–txʷ kʷi sčəlus

desire–ECS REM fish.tips

‘he wanted fish tips’

[AJ Basket Ogress, line 26]

b. x%aƛ’txʷ əlgʷəʔ kʷi səsq’ʷuʔləpəxʷ

x%aƛ’–txʷ əlgʷəʔ kʷi s=ʔəs–q’ʷuʔ=ləp=əxʷ

desire–ECS PL REM NM=STAT–gather=2PL.PO=now

‘they want you folks to gather’

(Hess 2006: 73, line 737)

The remote determiner is also used in negative expressions (Section 8.5) to introduce items

whose existence is negated (218a), or hypothetical entities involved in negated events (218b):

(218) a. xʷiʔ kʷi gʷəbiac kʷi gʷəstab

xʷiʔ kʷi gʷə=biac kʷi gʷə=stab

NEG REM SBJ=meat REM SBJ=what

‘there is no meat or anything’

(Hess 1998: 88, line 274)

206

b. xʷiʔ gʷadsq’puc ʔə kʷi gʷəstab

xʷiʔ gʷə=ad=s=q’pu–t–s ʔə kʷi gʷə=stab

NEG SBJ=2SG.PO=NM=pay–ICS–1SG.OBJ PR REM SBJ=what

‘don’t pay me anything’

(Hess 2006: 30, line 191)

A similar pattern is seen in information questions (8.4.2) in which the Given portion of the

question is not particular entity (as it might be in a request for identification — i.e., ‘what is

that?’) or is not certain to exist:

(219) a. stab kʷi gʷəshuy čəɬ

stab kʷi gʷə=s=huy čəɬ

what REM SBJ=NM=de.done 1PL.SUB

‘what can we do?’

(Hess 2006: 18, line 156)

b. čadəxʷ kʷi sbəq’ʷaʔ

čad=əxʷ kʷi sbəq’ʷaʔ

where=now REM heron

where was Heron?’

(Hess 2006: 19, line 177)

c. stab kʷi gʷəcəxʷtəɬəɬ dxʷgʷəd dxʷʔal ti tucəxʷtul’ʔa

stab kʷi gʷə=d=dəxʷ=təɬəɬ dxʷgʷəd dxʷ–ʔal ti

what REM SBJ=1SG.PO=ADNM=arrive.safely downward CNTRPT–at SPEC

tu=d=dəxʷ=tul’–ʔa

PAST=1SG.PO=ADNM=CNTRFG–be.there

‘what way can I get down safely to where I am from?’

(Hess 2006: 28, line 144)

The use of the remote determiner in desiderative, negative, and interrogative clauses is clearly

linked to the use of the cognate forms in other Salishan languages to express ‘non-assertion of

existence’ (Matthewson 1996), although in Lushootseed kʷi is also used in contexts where it

refers to entities that are known and asserted to exist (as in 212c above).

In addition to being used to localize and individuate entities in physical space, the

demonstrative determiners are also used to ground entities in discourse by linking particular

event-participants to “virtual” locations whose metaphorical proximity to the speaker is

207

proportional to their topicality and discourse salience. Consider, for example, the following

stretch of narrative:

(220) a. kʷədalcəxʷ tiʔəʔ kikəwič

kʷəd•alc=əxʷ tiʔəʔ ki–kəwič

take•arm=now PROX ATTN–hunchback

‘Little Hunchback grabs a limb’

b. sət’cut tul’ʔal tiʔəʔ xʷʔax%ʷaʔad

sət’–t–sut tul’–ʔal tiʔəʔ xʷʔax%ʷaʔad

lifted–ICS–REFL CNTRFG–at PROX basket

‘he lifts himself out of the clam basket’

c. ʔaˑ gʷəl, ʔux%ʷ

ʔa gʷəl ʔux%ʷ

INTJ SCONJ go

‘and then she goes on’

d. x%ʷul’əxʷ ʔuʔux%ʷ tsiʔiɬ ʔax%ʷadus

x%ʷul’əxʷ ʔu–ʔux%ʷ tsiʔiɬ ʔax%ʷadus

only=now PFV–go DIST Basket.Ogress

‘Basket Ogress just went on’

e. huy, təlawiləxʷ

huy təlawil=əxʷ

SCONJ run=now

‘and then he runs’

f. bəlkʷaxʷ tiʔəʔ kikəwič dxʷʔal tiʔiɬ q’il’bids

bəlkʷaxʷ tiʔəʔ ki–kəwič dxʷ–ʔal tiʔiɬ q’il’bids

return=now PROX ATTN–hunchback CNTRPT–at DIST canoe

‘Little Hunchback returned to his canoe’

[MS Basket Ogress, lines 39 – 44]

This stretch of discourse begins at a point in the narrative where the main character, kikəwič

‘Little Hunchback’, and his plight, being carried off by the Basket Ogress in her big clam basket,

are topical, and the narrator signals this topicality through the use of the proximal determiner in

(220a) and (b). The next two lines see a slight shift in narrative focus as the narrator describes

the actions of Basket Ogress (who continues on her way, unaware that Little Hunchback has

escaped her basket); however, the use of the distal determiner here maintains a certain narrative

distance from the Basket Ogress, who has not become the central figure of this discourse

208

episode. The central figure remains Little Hunchback, who appears again with the proximal

determiner in the last line, (220f). In a similar vein, the remote determiner can also be used in

this manner (most frequently to introduce backgrounded information), and a skilled narrator can

and does make heavy use of these three determiners, as well as the specific non-demonstrative

determiner ti, to control discourse and encode nuances of communicative structure. For this

reason, it is often difficult to make hard and fast predications about the use of the demonstrative

determiners or account for their particular uses in specific sentences drawn from texts; the

patterns presented above are the most prevalent and generalizable patterns, but the finer details

of the system are in need of further investigation.

Unlike the demonstrative determiners, the non-demonstrative determiners do not localize

entities in physical or discourse space, although, as specific determiners, they do single out

individuals from the type or set named by the expression they introduce. There are two members

of this group, the “plain” specific determiner, ti/tsi, and the unique determiner, tiʔacəc/tsiʔacəc.

The former is the more straightforward of the two, serving only to assert that the referent of an

NP or referring expression is a specific individual whose identity is Given or known from

discourse, as in (221):

(221) a. ɬild čəɬ ti dəgʷi tiʔiɬ dəxʷuwiliq’ʷid čəɬ

ɬil–d čəɬ ti dəgʷi tiʔiɬ dəxʷ=ʔu–wiliq’ʷi–d čəɬ

give.food–ICS 1PL.PO SPEC you DIST ADNM=PFV–ask–ICS 1SG.PO

‘we are giving you food, which is why we questioned you’

(Hess 1998: 80, line 69)

b. gʷəl ʔux%ʷ, ʔibəš ti sgʷəlub

gʷəl ʔux%ʷ ʔibəš ti sgʷəlub

SCONJ go travel SPEC pheasant

‘and then he goes, Pheasant travels’

(Hess 1998: 81, line 100)

209

The example in (221a) shows the use of the specific determiner with the second-person pronoun

dəgʷi ‘you’, which is inherently both Given and individualized.

In (221b), the definite

determiner introduces an NP that re-asserts the identity of the current discourse topic. In neither

instance is localization in physical or discourse space necessary for the identification of the

referent.

The second non-demonstrative determiner, tiʔacəc/tsiʔacəc, encodes specificity and an

additional semantic feature that, for lack of a better term, we will call uniqueness. Like ti/tsi, the

specific determiner asserts that the referent of an NP or referring expression is an individual of a

particular type whose identity is Given or known from discourse. Additionally, the unique

determiner emphasizes that the referent of the expression it introduces is that individual and no

other (cf. English expressions such as the very (one) or the one and only). Although the unique

determiner is not well-attested in texts, those environments where it does surface fall roughly

into two categories. The first of these are circumstances where it is used to introduce expressions

referring to previously-identified individuals that are central to the narrative and are known or

Given, but which are not topical at that particular moment in discourse:

(222) a. day’əxʷ haʔɬ tiʔəʔ səsq’ʷəlb ʔə tiʔacəc bəščəb

day’=əxʷ haʔɬ tiʔəʔ s=ʔəs–q’ʷəl–b ʔə tiʔacəc bəščəb

only=now good PROX NM=STAT–cooked–MD PR UNQ mink

‘what this Mink is roasting sure looks good’

(Hess 1998: 66, line 42)

b. diɬ tiʔacəc ƛ’ucutəb stiqtiqayuʔ tiʔəʔ ʔubəčalq

diɬ tiʔacəc ƛ’u=cut–t–əb stiq–tiqayuʔ tiʔəʔ ʔu–bəč•alq

FOC UNQ HAB=say–ICS–PASS DSTR–wolf PROX PFV–fallen•game

‘those who brought down the game [are] these very ones that are called wolves’

(Hess 1998: 73, line 202)

As discussed in Section 2.7.1, personal pronouns do not always require determiners; however, when they do

appear with determiners, second-person pronouns always take the specific determiner. First-person pronouns are not

attested with determiners in the current corpus.

210

(222a) comes from a context in which Changer, one of the protagonists of the story, has come

upon a salmon being roasted by Mink, who up until the beginning of the current episode had

been the main protagonist. By using the unique determiner, the narrator is underlining the fact

that this is the same Mink, the topic of the previous part of the story. In (222b), the unique

determiner serves to emphasize that it is indeed the wolves, and nothing else, that are responsible

for that kill, a fact that comes into play in the subsequent narrative.

In a similar vein to the example in (222b), the unique determiner is also quite frequently used

to introduce new discourse topics, as in the following stretch of narrative:

(223) a. hay gʷəl ʔux%ʷəxʷ tiʔiɬ cədiɬ dukʷibəɬ

hay gʷəl ʔux%ʷ=əxʷ tiʔiɬ cədiɬ dukʷibəɬ

SCONJ SCONJ go=now DIST he Changer

‘and so Changer goes’

b. ʔiˑˑ, bəʔəy’dxʷ tsiʔacəc k’aʔk’aʔ ʔi tiʔəʔ ʔalš

ʔi bə=ʔəy’dxʷ tsiʔacəc k’aʔk’aʔ ʔi tiʔəʔ ʔalš–s

INTJ ADD=find UNQ crow and PROX cross-sex.sibling–3PO

‘indeed, he finds Crow and her brother’

c. gʷəhaw’ə ʔuq’ʷic’il tsiʔəʔ cədiɬ k’aʔk’aʔ

gʷəhaw’ə ʔu–q’ʷic’il tsiʔəʔ cədiɬ k’aʔk’aʔ

semingly PFV–be.widowed PROX:FEM she crow

‘it seems that Crow had become a widow’

(Hess 1998: 70, line 131 - 133)

The first two lines of (223) are the last part of a stretch of discourse in which Changer is the

topic. In (223b), a new character, Crow, is introduced, and Crow becomes the topic of the

subsequent episode that begins at line (223c). The use of the unique determiner here both

highlights the introduction of a central but non-topical character and serves to link that character

to a personality well-known to the audience (Crow being a familiar figure from this and other

traditional stories) and whose (well-known) characteristics are to become central to the unfolding

narrative. In at least one case, this strategy is used at the very beginning of a story:

211

(224) ʔəsɬaɬlil tiʔiɬ ʔaciɬtalbixʷ dəxʷʔa ʔə tiʔacəc sbiaw

ʔəs–ɬaɬlil tiʔiɬ ʔaciɬtalbixʷ dəxʷ=ʔa ʔə tiʔacəc sbiaw

STAT–live DIST person ADNM=be.there PR UNQ coyote

‘people were living where Coyote was’

(Hess 1998: 91, line 1)

Although there can have been no previous mention of Coyote, he would have been a familiar

figure to a Lushootseed audience, who would have known all about Coyote and those personality

traits that identify him and drive the events in the story that is about to be told.

Opposed to the set of specific determiners are the two non-specific determiners, the non-

feminine tə and the feminine tsə. These are used to indicate that the following expression does

not refer to a particular individual, but instead refers to a type of individual, event, or

circumstance. Consider the examples in (225):

(225) a. huy gʷəl ǰuʔiləxʷ tə ʔaciɬtalbixʷ

huy gʷəl ǰuʔ–il=əxʷ tə ʔaciɬtalbixʷ

SCONJ SCONJ enjoy–INCH=now NSPEC person

‘and then the people had a good time’

(Hess 1995: 142, line 49)

b. ƛ’əƛ’iq’šəd tə ƛ’uʔuˑˑx%ʷ

ƛ’əƛ’iq’šəd tə ƛ’u=ʔux%ʷ

sapsucker NSPEC HAB=go

‘the one to go would be Sapsucker’

(Hess 2006: 18, line 164)

c. ʔuʔəy’ʔəy’dxʷ čəd tə sʔubədiʔ əlgʷəʔ ʔal kʷədiʔ t’aq’t čəda ʔuɬiltəb

ʔu–ʔəy’–ʔəy’dxʷ čəd tə sʔubədiʔ əlgʷəʔ ʔal kʷədiʔ t’aq’t

PFV–DSTR–find–ICS–DC 1SG.SUB NSPEC hunter PL PR REM.DMA inland

čəda ʔu–ɬil–t–əb

1SG.COORD PFV–give.food–ICS–PASS

‘I met some hunters up in the mountains and they gave it to me’

(Hess 1998: 82, line 147)

The non-specific expression in (225a), tə ʔaciɬtalbixʷ ‘the people’, does not refer to a specific

group of individuals so much as to an undefined set of people (in this story, those in the vicinity

of a beached whale). Similarly, the headless relative clause (Section 7.4.1) in (225b) identifies a

type of person (‘someone who will go’). In (225c), the speaker refers to a type of person,

212

hunters, rather than to a specific individuals whose identity is known to the addressee. Note that,

as shown by this example, the notion of non-specificity is relative to the proposition or the

utterance, rather than being absolute or absolutely dependent on the knowledge of the speaker: in

(225c), the speaker actually met the hunters and can identify them as individual members

belonging to the set of those who hunt. However, the addressee can not and the speaker’s choice

of determiner reflects accommodation to the common ground of shared knowledge (cf. Montler

2007).

In much the same vein, the non-specific determiner can be used to introduce expressions that

explicitly identify types of things, rather than referring to individuals:

(226) tuƛ’iq’ ʔə tsiʔəʔ čəgʷas tə dəxʷəšuucləp tə sx%ay’us ʔə ʔaciɬtalbixʷ ʔal tudiʔ sɬukʷalb

tu=ƛ’iq’ ʔə tsiʔəʔ čəgʷas–s tə dəxʷ=ʔəs–šuu–c=ləp

PAST=be.stuck PR PROX:FEM wife–3PO NSPEC ADNM=STAT–see–ALTV=2PL.PO

tə sx%ay’us ʔə ʔaciɬtalbixʷ ʔal tudiʔ sɬukʷalb

NSPEC head PR person PR DIST.DMA moon

‘[Moon] stuck to his wife, which is why you guys see a person’s head in the moon’

[HM Star Child, line 189]

The non-specific NP here, tə sx&ay’us ʔə ʔaciɬtalbixʷ ‘the head of a person’, refers to the

figurative image that Lushootseed culture projects onto the face of the full moon and as such is

not a reference to a particular human head, but rather a description of the type of shape that one

can see.

The non-specific determiner is also used for general locations, times, and non-unique or

repetitive events:

(227) a. tsiʔəʔ waq’waq’ ƛ’uʔal tə sc’p’alič

tsiʔəʔ waq’waq’ ƛ’u=ʔal tə sc’p’alič

PROX:FEM frog HAB=at NSPEC swamp

‘Frog would be in the swamp’

(Hess 2006: 3, line 3)

In terms of the story being told, this sentence is misspoken — what actually happened was that Moon’s wife,

Frog, stuck to his face, which is why the Lushootseed see the figure of a frog on the moon. Grammatically, however,

the sentence is fine.

213

b. qədəbtubəxʷ ʔə tiʔəʔ xʷiʔ ləhaʔɬ č’ətx% ʔal tə sxʷiʔ ʔə tiʔiɬ cədiɬ

qədəb–txʷ–b=əxʷ ʔə tiʔəʔ xʷiʔ lə=haʔɬ č’ətx% ʔal tə

fornicate–ECS–PASS=now PR PROX NEG NEGP=good kingfisher at NSPEC

s=xʷiʔ ʔə tiʔiɬ cədiɬ

NM=NEG PR DIST he

‘this no-good Kingfisher had sex with her while he [her husband] wasn’t home’

(Hess 2006: 12, line 45)

c. šuhu dəxʷdiɬəxʷ tiʔəʔ sɬukʷalb tə ləʔibəš ʔə tiʔəʔ diʔəʔ …

šuɬ ʔu dəxʷ=diɬ=əxʷ tiʔəʔ sɬukʷalb tə ləʔibəš ʔə tiʔəʔ diʔəʔ

see INT ADNM=FOC=now PROX moon NSPEC PROG=travel PR PROX here

‘so do you see? that is why [it is] the moon that is traveling by here’

[MW Star Child, line 117]

The first sentence here is a statement about where frogs are generally found — “in the swamp”

— and identifies a particular type of setting rather than a particular swamp. In (227b), a general

temporal setting expresses the time of a repeated or habitual action. The non-specific phrase in

(227c), a relative clause modifying sɬukʷalb ‘moon’, describes its head in terms of a repeated

event which happens every night that the moon is visible, rather than being a unique event that

can be localized at a fixed point in time.

In a somewhat more abstract use, the non-specific determiners turn up in adverbial

expressions of result or cause when these are general circumstances rather than specific events:

(228) a. gʷəl huyic ʔə tə cəxʷx%ʷil’

gʷəl huy–yi–d–s ʔə tə d=dəxʷ=x%ʷil’

SCONJ be done–DAT–ICS–1SG.OBJ PR NSPEC 1SG.PO=ADNM=be.lost

‘he did this to me, causing me to be lost’

(Hess 2006: 28, line 140)

b. ʔaciɬtalbixʷaladxʷ tə dəxʷtəšaləp

ʔaciɬtalbixʷaladxʷ tə dəxʷ=təša=ləp

human NSPEC ADNM=be.misfortune=2PL.PO

‘the [reason for] your misfortune [is] humans’

(Hess 2006: 76, line 820)

As noted in Hess (2006: 62, fn. 155), the expected form of the last word in this example is dəxʷtəšlap. The

variant pronunciation may be because this line is chanted on the tape, or because this is the specialized, accented

speech of birds.

214

In each of these cases, the non-specific expressions — all clauses nominalized with the

adjunctive nominalizer dəxʷ= (Section 7.4.2.2) — name a type of circumstance or state of affairs

that is either the result of the event expressed in the main clause (228a), or is caused by that

event (228b). Neither non-specific NP identifies a particular event.

In addition to these fairly straightforward uses, the non-specific determiners are also found in

other contexts which are related to, but perhaps not entirely predictable from, their basic meaning

of designating a type rather than a specific individual. One of these is their frequent appearance

in possessive constructions like those in (229):

(229) a. stabəxʷ tiʔəʔ ʔəsqʷəq’ʷil suʔəɬəd ʔə tə bədbədaʔ ʔə tiʔiɬ sgʷəlub

stab=əxʷ tiʔəʔ ʔəs–qʷəq’ʷil s=ʔu–ʔəɬəd ʔə tə bəd–bədaʔ

what=now PROX STAT–whitish NM=PFV–eat PR NSPEC DSTR–offspring

ʔə tiʔiɬ sgʷəlub

PR DIST pheasant

‘what is this whitish food of Pheasant’s children?’

(Hess 1998: 83, line 144)

b. haˑy, huy tə spigʷəd ʔə tsi x%ənimulicaʔ

hay huy tə spigʷəd ʔə tsi x%ənimulicaʔ

SCONJ be done NSPEC power.ceremony PR NSPEC:FEM X

ənimulicaʔ

‘then X

ənimulicaʔ’s spirit-power ceremony is done [i.e., finished]’

(Hess 1998: 59, line 78)

At first glance, these examples seem to be inconsistent with notion of tə as an non-specific

marker, since possessed entities such as those in (229) are highly specific and individuated.

However, as noted by Langacker (1987), one of the functions of possessive constructions

themselves is to identify individual entities by locating them with respect to another entity, their

possessor. In each of the examples in (229), the possessed is uniquely identified by its

association with a particular individual that is salient to discourse. In (229a), for instance, the

speaker (Raven) is curious about the sudden good fortune of his neighbour, Pheasant, and sends

his own children to investigate what it is Pheasant has given his children to eat. Thus, the

children are uniquely identified by their association with Pheasant and the NP tə bədbədaʔ ʔə

215

tiʔiɬ sgʷəlub ‘the children of Pheasant’ is grounded by the possessive phrase rather than by its

determiner. The same is true of the subject NP of (229b), tə spigʷəd ʔə tsi x&ənimulicaʔ

‘X

ənimulicaʔ’s spirit-power ceremony’, where the particular ceremony being referred to is

identified with reference to the main character of the story, X

ənimulicaʔ (a traditional name

given to Crow), rather than being grounded at a particular place or time through the use of one of

the specific determiners. In cases like these it seems that the non-specific determiner is chosen

not because the nominal it introduces is non-individuated or non-grounded, but rather because it

is grounded by something other than the determiner itself — specifically, by its possessor.

Of course, not all possessive expressions are necessarily grounded by their possessor. In

certain cases, for instance, the possessed is either more readily central to the discourse or better

known to the addressee, as in the example in (230):

(230) gʷəl ʔux%ʷ dxʷsəsaʔliʔ dxʷsxʷiʔxʷiʔxʷiʔ ʔalalš ʔə tsiʔəʔ sɬadəyʔ, čəgʷas ʔə tə luƛ’

dxʷsp’ayəq

gʷəl ʔux%ʷ dxʷ–sə–saʔliʔ dxʷs–xʷiʔ–xʷiʔxʷiʔ ʔal–alš ʔə tsiʔəʔ

SCONJ go CTD–PL–two PROC–DSTR–hunt PL–cross.sex.sibling PR PROX:FEM

sɬadəyʔ čəgʷas ʔə tə luƛ’ dxʷs–p’ayəq

woman wife PR NSPEC old PROC–carve.canoe

‘and they go, two hunters, brothers of the woman, wife of the old canoe-maker, ’

(Hess 2006: 50, line 206)

In the case of (230), the old canoe-maker is not non-specific in the sense of being non-

individuated. Quite the contrary, the old canoe-maker is a central figure in the story who has set

the trap into which the hunters are about to fall. However, in the context of this sentence, which

is about the hunters who are identified as the brothers of the woman, the fact that the woman is

the old man’s wife is offered as background information contained in an appositive phrase, rather

than as information intended to identify the woman and ground her in discourse.

Likewise, in (231), the possessor is not a particularly “localizable” entity and is introduced

by the non-specific determiner, in spite of its inherently specific (unique) nature:

216

(231) … čxʷa ɬuʔux%ʷ dxʷʔal kʷi ɬusbəčax%ad ʔə tə swatixʷtəd …

čxʷa ɬu=ʔux%ʷ dxʷ–ʔal kʷi ɬu=sbəčax%ad ʔə tə swatixʷtəd

2SG.COORD IRR=go CNTRPT–at REM IRR=edge PR NSPEC land

‘… and you will go to the edge of the world …’

(Hess 2006: 51, line 220)

Here, the non-specific possessor, tə swatixʷtəd ‘the world’ (lit. ‘the land’), does not serve to

locate the event in any meaningful way, ‘the world’ being a rather broad region, whereas ‘the

edge’ is a more practical reference point for the event (and so is localized in remote physical

space by kʷi). Once again, it seems that the non-specific determiner is chosen for an embedded

NP when the nominal it introduces does not ground the matrix NP in which they are contained.

The non-specific determiners also turn up in a few contexts where their import is not entirely

clear. The most frequent of these is in headless relative clauses that identify their referents in

terms of undesirable or negative characteristics, as in these examples:

(232) a. ləluƛ’iləxʷ tə sʔušəbabdxʷ stutədəq

lə=luƛ’–il=əxʷ tə sʔušəbabdxʷ stu–tədəq

PROG=old–INCH=now NSPEC poor.guy ATTN–slave

‘the poor little slave is growing up’

[HM Star Child, line 144]

b. bək’ʷək’ʷ čəɬ ɬux%ədəd čɬa ɬuhudabacəd tsə xʷiʔ ləhaʔɬ

bək’ʷ–ək’ʷ čəɬ ɬu=x%əd–əd čɬa ɬu=hud•abac–əd tsə

INTNS–all 1PL.SUB IRR=push–ICS 1PL.COORD IRR=fire•body–ICS NSPEC:FEM

xʷiʔ lə=haʔɬ

NEG NEGP=good

‘every one of us will push her [into the fire] and burn up the body of the no-good [one]’

[DM Basket Ogress, line 60]

In both sentences, the referent of the relative clause is known to both speaker and addressee and

is, in fact, highly topical in the context (perhaps making it possible to allude to the entity more

indirectly without properly grounding it in the first place). The use of the non-specific determiner

is not obligatory in these circumstances, as shown by the example in (233):

217

(233) ʔuˑ tiʔəʔ xʷiʔ ləhaʔɬ

ʔu tiʔəʔ xʷiʔ lə=haʔɬ

INTJ PROX NEG NEGP=good

‘oh, the no-good [one]!’

(Hess 1998: 71, line 157)

This suggests that the choice of determiner may have some stylistic effect. The fact that it is

associated with negative characteristics could also suggest that it is a (im)politeness strategy,

although this correlation may also be the result of a gap in the corpus. Pending the discovery of

further examples, the issue will have to be left for future investigation.

As well as being inflected for a wide variety of deictic categories, determiners are inflected

for a natural gender, and two of them show quasi-inflectional marking for number as well. The

basic gender distinction is between non-feminine/unmarked and feminine. Historically, this may

have been implemented by means of an infix *-s- occurring immediately after the first consonant

of the determiner (ti ‘SPECIFIC’ vs. tsi ‘SPECIFIC FEMININE’, kʷi ‘REMOTE’ vs. kʷsi ‘REMOTE

FEMININE’), a hypothesis that is reflected in Lushootseed orthography, which writes <ts> rather

than <c> for the feminine determiners. The feminine forms of determiners are used obligatorily

when introducing expressions referring to biologically female humans and female personified

animals in traditional stories:

(234) a. ɬadəxʷɬčiltxʷ dxʷʔal kʷsi adčəgʷas ʔi kʷi adbədbədaʔ

ɬu=ad=dəxʷ=ɬčil–txʷ dxʷʔal kʷsi ad–č əgʷas ʔi kʷi

IRR=2SG.PO=ADNM=arrive–ECS PR REM:FEM 2SG.PO–wife and REM

ad–bəd–bədaʔ

2SG.PO–DSTR–child

‘so that you can bring it to your wife and children’

(Hess 1998: 86, line 237)

b. ƛ’ub čəxʷ ʔubiqʷyitəb ʔə tsiʔacəc tubədaʔs

ƛ’ub čəxʷ ʔu–biqʷ–yi–t–əb ʔə tsiʔacəc tu=bədaʔ–s

okay 2SG.SUB PFV–loosen–DAT–ICS–PASS PR UNQ:FEM PAST=child–3PO

‘you should be permitted his former daughter’

(Hess 1998: 98, line 204)

218

While gender is obligatorily indicated for expressions with singular human referents, the gender

distinction is neutralized in the plural: the non-feminine form of the determiner is used for all

plural referents, even when one or all of them are female:

(235) a. hay čxʷa ɬukʷədxʷ tiʔiɬ čaagʷəs

hay čxʷa ɬu=kʷəd–dxʷ tiʔiɬ čaagʷəs–s

SCONJ 2SG.COORD IRR=taken–DC DIST wife:PL–3PO

‘and then you can take his wives’

(Hess 2006: 23, line 34)

b. ləqdubəxʷ ʔə tiʔəʔ sɬəɬadəyʔ legʷax%ʷ

ləq–dxʷ–b=əxʷ ʔə tiʔəʔ sɬə–ɬadəyʔ le–gʷax%ʷ

hear–DC–PASS=now PR PROX PL–woman PROG=walk

‘she was heard by the women who were walking about’

[MW Star Child, line 21]

This pattern holds even in contexts where the plurality of the referent is not marked

grammatically (as it is in the examples in 235), but is to be inferred solely from context. As

shown in (236), gender is also marked on the determiners when these are used with pronouns,

including the second-person singular pronoun, dəgʷi.

(236) xʷiʔ gʷəsuɬaldxʷ čəɬ tsi dəgʷi

xʷiʔ gʷə=s=ʔu–ɬal–dxʷ čəɬ tsi dəgʷi

NEG SBJ=NM=PFV–remove.from.fire–DC 1PL.SUB SPEC:FEM you

‘we can’t get you out of the fire’

[LA Basket Ogress, line 87]

However, with pronouns gender-marking seems to be optional, and second-person pronouns with

non-feminine determiners and female antecedents are found in texts:

(237) tux%ʷ čəɬ ləcuhiiɬbid ti dəgʷi ʔə ti adsux%ʷuyilcəb

tux%ʷ čəɬ ləcu–hiiɬ–bi–d ti dəgʷi ʔə ti

just 1PL.SUB CONT–happy–MAP-ICS SPEC you PR SPEC

ad=s=ʔu–x%ʷuy•ilc–əb

2SG.PO=NM=PFV–steam.cook•round.object–MD

‘we are just happy for you as you are steam-cooking’

[AJ Basket Ogress, line 83]

There are no attestations of the first-person singular pronoun, ʔəca, with a determiner in the corpus, so it is

impossible to say whether or not a feminine determiner would be used by a female speaker in this case as well.

219

This sentence is drawn from a text where the speakers are addressing a female character, Basket

Ogress, who is consistently referred to in the third-person with the feminine forms of the

determiners, both by the narrator and by the characters speaking in (237).

With words that refer to humans, gender-marking on the determiner is often the only way to

make distinctions of sex that are encoded lexically in other languages:

(238) a. i. gʷat əw’ə tsiʔiɬ č’ač’as ʔal tudiʔ

gʷat əw’ə tsiʔiɬ č’ač’as ʔal tudiʔ

who PTCL DIST:FEM child at DIST.DMA

‘who is that girl over there?’

ii. diɬ dbədaʔ tsiʔiɬ

diɬ d–bədaʔ tsiʔiɬ

FOC 1SG.PO–offspring DIST:FEM

‘she is my daughter’ [lit. ‘that

FEM

is my offspring’]

(Hess & Hilbert 1976: I, 10, ex. 4)

b. i. gʷat əw’ə tiʔiɬ č’ač’as ʔal tudiʔ

gʷat əw’ə tiʔiɬ č’ač’as ʔal tudiʔ

who PTCL DIST child at DIST.DMA

‘who is that boy/child over there?’

ii. diɬ dbədaʔ tiʔiɬ

diɬ d–bədaʔ tiʔiɬ

FOC 1SG.PO–offspring DIST

‘that is my son’ [lit. ‘that

MASC

is my offspring’]

(Hess & Hilbert 1976: I, 10, ex. 3)

Lushootseed does not have any specific words corresponding to the English words boy and girl

or son and daughter — instead, it makes use of two gender-neutral terms, č’ač’as ‘child’ and

bədaʔ ‘offspring’. In order to refer specifically to a female child, the feminine form of the

determiner is required, as in (238a). When referring to a male, the non-feminine form is used, as

in (238b) — note, however, that the same form would be used if the gender of the referent were

unknown by the speaker. As shown by the responses to the questions in (238), gender inflection

applies to the pronominal uses of the determiners as well.

220

The feminine forms of the deictics are also used to refer to small things, often with a

connotation of cuteness or affection (cf. the use of diminutives in this way in Spanish and other

Romance languages):

(239) a. tiləb ləkʷəd[ad] tsiʔəʔ qʷiqʷɬayʔulč

tiləb lə=kʷəda–d tsiʔəʔ qʷi–qʷɬayʔ•ulč

suddenly PROG=take–ICS PROX:FEM ATTN–wooden.dish•dish

‘right away they took this little wooden dish’

(Hess 2006: 38, line 407)

b. gʷəl ʔal tudiʔ ʔal kʷsi little creek t’aq’tylus ʔə tiʔəʔ ʔal Clear Lake

gʷəl ʔal tudiʔ ʔal kʷsi little creek t’aq’t•y•alus ʔə

SCONJ at DIST.DMA at REM:FEM little creek landward•CNN•eye PR

tiʔəʔ ʔal Clear Lake

PROX at Clear Lake

‘and it is there at the little creek at the upper end of Clear Lake there’

[DS Star Child, line 66]

This use of the feminine forms of determiners depends entirely on the intent of the speaker: small

objects can be referred to with the plain determiners, just as large objects (such as hunting

canoes) can be referred to with the feminine determiners. The use of the feminine determiners to

express diminutivity is independent of the use of attenuative Type II reduplication (Section 5.2),

although the two show considerable overlap in meaning.

In addition to being inflected for gender when used to introduce NPs, determiners are also

inflected for gender when they introduce headless relative clauses (Section 7.4.1), as in (240):

(240) a. xʷiʔ gʷəbəslək’ʷtəbs ʔə tsiʔəʔ cədiɬ ʔəsx%əɬ

xʷiʔ gʷə=bə=s=lək’ʷ–t–əb=s ʔə tsiʔəʔ cədiɬ ʔəs–x%əɬ

NEG SBJ=ADD=NM=eat–ICS–PASS=3PO PR PROX:FEM he STAT–sick

‘this sick one would not eat them’

(Hess 2006: 14, line 71)

b. šəq tsiʔiɬ səscaq’abactubs

šəq tsiʔiɬ s=ʔəs–caq’•abac–txʷ–b=s

be.high DIST:FEM NM=STAT–spear•body–ECS–PASS=3PO

‘she who is impaled is up there’

(Hess 2006: 18, line 154)

221

c. tuqiʔqəl’adiʔ kʷsi tuhəliʔ

tu=qiʔ–qəl’adiʔ kʷsi tu=həliʔ

PAST=ATTN–Qəl’adiʔ REM:FEM PAST=be.alive

‘the one who was alive was Qəl’adiʔ’

[ML Basket Ogress, line 216]

In these examples, the relative clause is subject-centred — that is, it refers to the subject of the

embedded clause — which in all of these cases is a female, thereby requiring the feminine form

of the determiner. The same phenomenon is observed with diminutive expressions:

(241) x%ʷul’əxʷ tuk’ʷik’ʷəd kʷsi tuləskʷəd əlgʷəʔ ʔə tiʔiɬ haʔɬ sʔuləx% …

x%ʷul’=əxʷ tu=k’ʷi–k’ʷəd kʷsi tu=ləs–kʷəd əlgʷəʔ ʔə

only=now PAST=ATTN–how.many REM:FEM PAST=PROG.STAT–taken PL PR

tiʔiɬ haʔɬ sʔuləx%

DIST good dentalia

‘only a small amount remained to them of those many good dentalia’

(Hess 2006: 70, line 669)

In this example, the feminine determiner kʷsi is used to emphasize the small amount of the

dentalia (a valuable trade item) left to the characters in a story after an unfortunate accident.

Similarly, the feminine determiner is also used with diminutives in constructions with

nominalized clauses, such as that in (242):

(242) diɬ tsiʔəʔ ƛ’ustab tsiʔəʔ dəxʷəʔatəbəds əlgʷəʔ

diɬ tsiʔəʔ ƛ’u=stab tsiʔəʔ dəxʷ=lə=ʔatəbəd=s əlgʷəʔ

FOC PROX:FEM HAB=stab PROX:FEM ADNM=PROG=die=3PO PL

‘it is this that is the reason that they are dying’

(Hess 2006: 63, line 522)

This sentence contains a nominalization, dəxʷəʔatəbəds əlgʷəʔ ‘the reason they are dying’,

introduced by the feminine determiner tsiʔəʔ, used here because the cause of death in this case is

tiny feathers that have been flung at the victims (dwarves) by attacking ducks. However,

feminine forms of the determiners are also found introducing nominalized clauses where the

reference of the expression is not the female participant, as in the examples in (243):

222

(243) gʷəl ʔuxʷəcəd tul’ʔal tsiʔiɬ dəxʷəscaq’tubs

gʷəl ʔu–xʷəc–əd tul’–ʔal tsiʔiɬ dəxʷ=ʔəs–caq’–txʷ–b=s

SCONJ PFV–remove–ICS CNTRFG–at DIST:FEM ADNM=STAT–impaled–ECS–PASS=3PO

‘he took her off from where she had been impaled’

(Hess 2006: 20, line 200)

In (243), the adjunct nominalization (Section 7.4.2.2) dəxʷəscaq’tubs ‘where s/he had been

impaled’ is introduced by the feminine determiner tsiʔiɬ, despite the fact that the nominalization

expresses a location rather than the identity female entity. A similar pattern of “oblique” gender-

marking is also found in negative expressions (Section 8.6):

(244) a. diič’uʔ kʷsi xʷiʔ gʷəsʔəƛ’s

diič’uʔ kʷsi xʷiʔ gʷə=s=ʔəƛ’=s

one.person REM:FEM NEG SBJ=NM=come=3PO

‘that [female] one has not come’

[DS Star Child, line 272]

b. bəxʷiˑʔ kʷsi gʷəbəstab gʷəbəʔah

bə=xʷiʔ kʷsi gʷə=bə=stab gʷə=bə=ʔah

ADD=NEG REM:FEM SBJ=ADD=what SBJ=ADD=exist

‘again there was nothing that had been there’

(Hess 2006: 38, line 411)

The embedded clause in (244a), kʷsi xʷiʔ gʷəsʔəƛ’s ‘her not coming’, is a sentential nominal

(Section 7.4.2.1) whose reference is an entire event, rather than a headless relative clause whose

reference is an event-participant. Nevertheless, the fact that the subject of the embedded clause is

female triggers the use of the feminine form of the determiner (much as it triggers the feminine

possessive form her in the English gloss). In (244b), the embedded clause is also a sentential

nominal, kʷsi gʷəbəstab gʷəbəʔah ‘there being (something) there again’, which is negated by

the clausal predicate, xʷiʔ. The source of the feminine gender is the fact that “there” is a small,

pretty dish that has consistently been referred to as feminine in the preceding discourse. Thus, it

seems that the presence of a female/diminutive entity in an event is enough to cause its

expression to be marked as feminine.

Two of the demonstrative determiners — the proximal and the distal — have optional plural

forms:

David Beck 10-2-7 2:19 PM

Comment: this is an incorrect description of

these sentences. the subject NP is a subject

centred RC based in the impersonal negative

construction ‘(it) is not her coming’

223

(245) a. q’ʷuʔtəbəxʷ tiʔiʔiɬ siʔiʔab

q’ʷuʔ–t–əb=əxʷ tiʔ–iʔiɬ siʔ–iʔab

gather–ICS–PASS=now PL–DIST PL–noble

‘the high-ranking people were gathered’

(Hess 1995: 144, line 30)

b. day’ tiʔiɬ wiw’su tiʔiɬ səʔəɬəds tiʔiʔəʔ tusƛ’alqəb, sxʷəyuq’ʷ ʔi tsiʔəʔ cədiɬ ʔay’əds

day’ tiʔiɬ wiw’su tiʔiɬ s=lə=ʔəɬəd=s tiʔ–iʔəʔ tu=sƛ’alqəb

only DIST children DIST NM=PROG=feed.on=3PO PL–PROX PAST=monster

ʔi tsiʔəʔ cədiɬ ʔay’əd–s

and PROX:FEW he companion–3PO

‘what the monsters, Basket Ogress and her companion, were feeding on [were] just the

children’

[ML Basket Ogress, line 194]

c. ɬusʔəɬəds tiʔiʔəʔ stawixʷaʔɬ

ɬu=sʔəɬəd–s tiʔ–iʔəʔ stawixʷaʔɬ

IRR=food–3PO PL–PROX children

‘the children will be her food’

[MS Basket Ogress, line 28]

In (245a), the plural form of the distal determiner is shown introducing a noun that has been

marked for plurality itself by a Type III plural reduplication (Section 5.3.4). As shown in (245b),

however, the fact that an NP is plural does not require that the noun itself be marked for

plurality, nor does the inherent plurality of a noun — like wiw’su ‘children’ — require the use of

a plural determiner (although a plural determiner can be used with such nouns, as in 245c). This

optionality of plural-marking indicates that, rather than being strictly inflectional, plural-marking

of determiners is quasi-inflectional (Mel’čuk 1993-2000, 2006). Note also that there are no

feminine plural forms of the determiners and, as seen in (245b), when a collective of females is

referred to, gender is neutralized.

2.5 Adverbs and other predicate-modifiers

2.5.1 Lexical adverbs and adverbial particles

Another small lexical class is the class of adverbs. These are words whose primary syntactic

function is as the modifier of the head of a predicate phrase, and they can be subdivided into two

224

types — lexical adverbs and adverbial particles. Lexical adverbs are potentially predicative

words and may appear in a variety of other syntactic roles, whereas adverbial particles function

exclusively as predicate modifiers. Adverbs are also distinguishable from adverbial particles in

that many of the former take (primarily verbal) derivational affixes and undergo reduplication,

whereas adverbial particles are more or less inert morphologically-speaking. In their role as

predicate-modifiers, however, both types of word show the same set of syntactic behaviours and

form a fairly obvious lexical class.

A list of lexical adverbs is given in Table 59:

bək’ʷ ‘all, completely’

hikʷ ‘big, very’

cəɬul’ ‘previously’

hiqab ‘too much’

cukʷ ‘only’ (Sk)

ƛ’al’ ‘also’

day’ ‘only, especially’ (Sn., SL)

ƛ’ub ‘well’

diɬ ‘just that one; FOCALIZER’

tux&ʷ ‘just, merely’

gʷəhaw’ə ‘apparently’

xʷɬub ‘in fact’

haʔkʷ ‘long time’

xʷiʔ ‘no, not’

haʔɬ ‘good’

xʷuʔələʔ ‘maybe’

həlaʔb/laʔb ‘really, a lot; truly’

yəhaw’/yaw’ ‘only if; not until’

Table 59: Lexical adverbs

As noted above, most of the words in Table 59 take at least some of the derivational affixes

particular to verbs.

√bək’ʷ ‘all, completely’

bək’ʷdxʷ

‘manage to get all of ’

bək’ʷil ‘get used up, be done exhaustively’

bək’ʷildxʷ

‘consume of all of ’

√cukʷ ‘only’

cukʷtxʷ ‘just allow ’

cugʷaɬ ‘be the last ’

√diɬ ‘just that one; FOCALIZER’

ʔəsdiɬgʷas ‘be the same way’

√gʷəhaw’ə ‘apparently’

gʷəhaw’ad ‘blurt out ’

√haʔɬ ‘good’

haʔləb ‘be nice [weather]’

haʔɬid ‘make  good’

haʔɬil ‘become good’

√hikʷ ‘big’

hikʷtxʷ ‘respect ’

higʷəd ‘uphold , support ’

higʷil ‘become noble’

higʷild ‘make  bigger’

√hiqab ‘too much’

hiqabil ‘become too much’

√ƛ’al’ ‘also’

ƛ’al’yib ‘add ’

√ƛ’ub ‘good, well’

ƛ’ubad

‘agree to ’

ƛ’ubil ‘get better’

ƛ’ubildxʷ

‘manage to improve ’

ƛ’ubtxʷ

‘have  fixed’

225

√xʷiʔ ‘no, not’

xʷəɬ ‘run out of ’

xʷiʔəd ‘refuse ’

xʷiʔil ‘become non-existent’

xʷiʔtxʷ ‘refuse to allow ’

√yəhaw’ ‘only if; not until’

yəhaw’txʷ ‘go a head with ’

Table 60: Verbs derived from lexical adverbs

The adverbs that do not seem to be attested as parts of derived forms are: cəɬul’ ‘previously’,

day’ ‘only, especially’, haʔkʷ ‘long time’, həlaʔb/laʔb ‘really, a lot’, tux&ʷ ‘unexpectedly’, xʷɬub

‘in fact’, and xʷuʔələʔ ‘maybe’. In spite of not all having the derivational possibilities of verbs,

all of the words in Table 59 can potentially function (where semantically appropriate) in either of

the principal syntactic roles of lexical adverbs — predicate-modifier and clausal predicate.

When used as predicate-modifiers, lexical adverbs generally appear pre-verbally, as the first

element in the clause, as in (246):

(246) a. haʔkʷ ƛ’uyabuk’ʷ

haʔkʷ ƛ’u=yabuk’ʷ

long.time HAB=fight

‘they had been fighting for a long time’

(Hess 2006: 75, line 786)

b. ʔuˑ, tux%ʷ čəɬ ʔəsq’ʷuʔ

ʔu tux%ʷ čəɬ ʔəs–q’ʷuʔ

INTJ just 1PL.SUB STAT–gathered

‘oh, we have just gathered here’

(Hess 1995: 147, line 7)

c. xʷuʔələʔ ƛ’ub čəd bəqadadid ʔə tiʔiɬ səsq’ʷəlbs čəda gʷəʔəɬəd

xʷuʔələʔ ƛ’ub čəd bə=qada–di–d ʔə tiʔiɬ s=ʔəs–q’ʷəl–b=s

maybe well 1SG.SUB ADD–steal–SS–ICS PR DIST NM=STAT–cooked–MD=3PO

čəda gʷə=ʔəɬəd

1SG.COORD SBJ=feed.on

‘I guess I ought to steal what he is roasting and eat it’

(Hess 1998: 67, line 45)

As shown in (246b) and (c), lexical adverbs attract matrix subject-markers and other sentence-

second particles (Sections 8.1.1 and 2.5.2, respectively). It is also possible to have more than a

single adverb in a clause (246c), their relative order — and their relative order with respect to

sentence-second particles — being determined by emphasis and stylistic considerations.

226

Although by far the most common position for lexical adverbs is pre-verbal, there are at least

a few instances where an adverb follows the verb:

(247) tubaɬatəbəxʷ xʷuʔələʔ ʔə tiʔəʔ ʔaciɬtalbixʷ

tu=baɬa–t–əb=əxʷ xʷuʔələʔ ʔə tiʔəʔ ʔaciɬtalbixʷ

PAST=shaman.cure–ICS–PASS=now maybe PR PROX person

‘perhaps the people treated her by shaman-curing’

(Hess 2006: 21, line 225)

This variation is probably stylistic, but the issue deserves further investigation.

Lexical adverbs can also function as the heads of predicate phrases:

(248) a. hikʷ tiʔəʔ ʔalʔals

hikʷ tiʔəʔ ʔalʔal–s

big PROX house–3PO

‘his house was really big’

[MS Basket Ogress, line 26]

b. ƛ’uhaʔkʷ kʷi ƛ’usgʷəšbads

ƛ’u=haʔkʷ kʷi ƛ’u=s=gʷəšbad=s

HAB=long.time REM REM=NM=disappear=3PO

‘for a long time he would disappear’ (lit. ‘his disappearance would be a long time’)

(Hess 2006: 56, line 350)

However, even when used in this role, adverbs are not inflected for aspect, and in this respect

they pattern with non-verbal predicates (see Section 8.3 below).

Adverbs used as predicate heads can also be modified by other adverbs or predicate particles

(Section 2.5.3), as in (249):

There is one exception to this generalization in the corpus used for this grammar:

(i) ʔəsƛ’ub ɬuhəliʔ

ʔəs–ƛ’ub ɬu=həliʔ

STAT–well IRR=alive

‘let it be alive!’

(Hess 2006: 48, line 159)

This sentence is also unusual in that imperative/exhortative expressions are generally not marked for aspect any

more than adverbs are. It may be that this is an idiomatic expression or lexicalization of ƛ’ub with the stative prefix,

or that ƛ’ub when used in this sense has taken on some of the inflectional properties of a verb.

227

(249) a. day’əxʷ haʔɬ tiʔəʔ səsq’ʷəlb ʔə tiʔacəc bəščəb

day’=əxʷ haʔɬ tiʔəʔ s=ʔəs–q’ʷəl–b ʔə tiʔacəc bəščəb

only=now good PROX NM=STAT–cooked–MD PR UNQ mink

‘what this Mink is roasting [is] really good’

(Hess 1998: 66, line 42)

b. diɬiɬ əw’ə tiʔəʔ saʔ luƛ’

diɬ–iɬ əw’ə tiʔəʔ saʔ luƛ’

INTNS–just.that.one PTCL PROX bad old

‘that very bad old man is the one’

(Hess 2006: 31, line 233)

In (249a), the modifying adverb day’ precedes the adverb haʔɬ acting as the clausal predicate,

while in (249b) the predicate particle əw’ə follows the (reduplicated) adverbial sentence

predicate diɬiɬ.

Meaning permitting, adverbs can also modify nouns, as in (250):

(250) a. ʔəsɬaɬlil kʷsi bšč’ad ʔal tə hikʷ ʔalʔal

ʔəs–ɬaɬlil kʷsi bšč’ad ʔal tə hikʷ ʔalʔal

STAT–live REM:FEM louse at NSPEC big house

‘Louse lives in a big house’

[EK Lady Louse, line 1]

b. tux%ʷ huy hikʷhikʷ haʔɬ sʔuləx% stabigʷs …

tux%ʷ huy hikʷ–hikʷ haʔɬ s=ʔuləxʷ stabigʷs

just SCONJ DSTR–big good NM=gathered possession

‘there were lots of good possessions to be gathered’

(Hess 2006: 60, line 437)

In this way, adverbs parallel verbs (see Sections 7.3 and 7.3). However, unlike verbs acting as

modifiers (as, for example, the s=nominal sʔuləx& ‘gathering’ in 250b), adverbs used in this way

tend not to follow the modified element, whereas verbs can either precede or follow their

modificand. Not unexpectedly, verbs used as modifiers can in turn be modified by adverbs:

(251) huy qa, hikʷ qa biac tiʔiɬ kʷagʷičəd

huy qa hikʷ qa biac tiʔiɬ kʷagʷičəd

SCONJ much big much meat DIST elk

‘well, it was a lot, this elk was really a lot of meat’

(Hess 1998: 84, line 183)

228

In this example, the adverb hikʷ ‘big’ modifies the verbal radical qa ‘be a lot’, which itself

modifies the noun biac ‘meat’. Just as verbs can not be used to modify other verbs, verbs can not

serve as modifiers of modifiers in constructions like that in (251).

The second sub-type of adverb is the adverbial particle. Adverbial particles differ from

lexical adverbs in two main respects. The first is that they can not function as the head of a

predicate phrase. The second is that they are morphosyntactically “inert” in the sense that they

are not inflected at the lexical or the phrasal level and do not undergo any kind of derivation. On

both counts, these words fit comfortably into the traditional category of particle. A list of

adverbial particles is given in Table 61:

cick’ʷ ‘very’ (NL)

lil ‘much, much more’

cay ‘very’ (SL)

put ‘very much, greatly’

ck’ʷaqid ‘always’

tiləb ‘immediately’

daʔxʷ/daw’ ‘just now’

x&əɬ ti ‘as though, like’

dəxʷ ‘??’

x&ʷul’ ‘just, only’

gʷaʔxʷ ‘eventually, soon’

Table 61: Adverbial particles

While most of these particles are frequent in the current analyzed textual corpus, the particle

dəxʷ is attested only in a single narrative from a Duwamish story-teller; its meaning is unknown.

The item lil in the table is a lexicalized use of the verb lil ‘be far’ which appears to be particular

to comparative constructions (Section 8.8), where it takes on a meaning as an intensifier of the

quality being compared:

(252) lil čəxʷ ʔiɬqʷiq’ʷ tul’ʔal ʔəca

lil čəxʷ ʔiɬ–qʷiq’ʷ tul’–ʔal ʔəca

more 2SG.SUB PRTV–strong CNTRFG–at I

‘you are a lot stronger than I’

(Bates, Hess & Hilbert 1994: 137)

While the metaphor is clear, this use of lil is hardly transparent, and so merits treatment as a

separate lexical item belonging to a separate part of speech.

David Beck 10-2-7 2:19 PM

Comment: syntax of XElh ti might be worth

a mention

229

huy gʷəl ƛ’uflop x%əɬ tihəxʷ

huy gʷəl ƛ’u=flop x%əɬ ti=həxʷ

SCONJ

HAB

=flop seemingly=now

‘And then it would sort of flop.’

SH 141

In terms of their distribution, adverbial particles appear exclusively as modifiers of the heads

of predicate phrases, as in (253):

(253) a. tiləb ʔuhaydxʷ ʔulək’ʷtəb tiʔiɬ ʔalʔalš ʔə tsi sxʷiyuk’ʷ

tiləb ʔu–haydxʷ ʔu–lək’ʷ–t–əb tiʔiɬ ʔal–ʔalš–s

immediately PFV–know PFV–eaten–ICS–PASS DIST DSTR–cross.sex.sibling–3PO

ʔə tsi sxʷiyuk’ʷ

PR SPEC:FEM ogress

‘right away he realized that his siblings were being eaten by the Basket Ogress’

[JS Basket Ogress, line 19]

b. put ʔəsluʔluʔ tiʔəʔ x%ʷubt

put ʔəs–luʔ–luʔ tiʔəʔ x%ʷubt

greatly STAT–DSTR–have.holes PROX paddle

‘the paddle was really full of holes’

[AJ Basket Ogress, line 69]

Like lexical adverbs, adverbial particles are pre-verbal and attract sentence-second particles

(Section 2.5.2), including the matrix subject-markers (Section 8.1.1), as in (254):

(254) a. cickʷ čəd ʔəxʷʔux%ʷəb

cickʷ čəd ʔəs–dxʷ–ʔux%ʷ–əb

very 1SG.SUB STAT–CTD–go–DSD

‘I very much want to go’

(Hess 1995: 90, ex. 21)

b. ck’ʷaqid sixʷ x%ʷul’ ʔuʔəɬəd tiʔəʔ qaw’qs

ck’ʷaqid sixʷ x%ʷul’ ʔu–ʔəɬəd tiʔəʔ qaw’qs

always PTCL only PFV–feed.on PROX raven

‘Raven is always just eating’

(Hess 1995: 90, ex. 24)

As seen in the second example, sequences of adverbial particles can occur, in which case all

appear before the verb and their relative ordering is determined by emphasis and stylistic

considerations. Adverbial particles can also co-occur with lexical adverbs, as in (255):

230

(255) a. cickʷ həlaʔb ʔəsqad tiʔiɬ stiqiw

cickʷ həlaʔb ʔəs–qad tiʔiɬ stiqiw

very really STAT–fast DIST horse

‘that horse is really very fast’

(Hess 1995: 90, ex. 15)

b. ƛ’ubəxʷ čəxʷ x%ʷul’əxʷ ɬubəščəb

ƛ’ub=əxʷ čəxʷ x%ʷul’=əxʷ ɬu=bəščəb

well=now 2SG.SUB only=now IRR=mink

‘it is better that you just become a mink’

(Hess 1998: 69, line 122)

In these cases, the relative ordering of the lexical adverb and the adverbial particle is free,

depending once again on style and emphasis. Adverbial particles can also be used to modify

lexical adverbs when these function as predicate heads:

(256) put hikʷ tiʔəʔ k’tuʔs

put hikʷ tiʔəʔ k’tuʔ–s

really big PROX belly–3PO

‘his belly was really big’

[MW Star Child, line 49]

The adverbial particle in such sentences is clause-initial, preceding the adverbial predicate.

2.5.2 Locative adverbs

In addition to the lexical adverbs discussed in Section 2.5.1, Lushootseed has a small,

separate class of locative adverbs. Several of these are given in Table 62:

ʔa ‘there’

gʷəd ‘down’

ʔaɬx&ad ‘downstream’

lil ‘far’

ʔilgʷiɬ ‘ashore, at the shoreline’

q’ixʷ ‘upstream’

čaʔkʷ ‘seaward’

šq ‘high, up in the air’

čad ‘where?’

t’aq’t ‘inland, landward’

č’it ‘nearby’

Table 62: Locative adverbs

Of these words, ʔa ‘there’ is a specialized use of the verb ʔa ‘be there’. The words in Table 62

include terms for marine and riverine orientation (the Salishan equivalent of North-South

Because the identification of words of this class depends crucially on their attestation in a particular syntactic

environment (post-predicate position — see below), it is not possible to be certain that this list is exhaustive, or

whether there are other words that also belong to this class but are simply not attested in post-predicate position in

the present corpus.

David Beck 10-2-7 2:19 PM

Comment: what about di? ‘other side’?

231

cardinal directions in European languages), terms for ‘nearby’/‘far’ and ‘up’/‘down’, and the

locative interrogative word čad ‘where?’ (Section 2.6.4). The primary characteristic of words of

this class that distinguishes them from other parts of speech and from other adverbs is their

ability to function as post-predicate (rather than pre-predicate) modifiers of verbs, as in (257):

(257) a. ƛ’upədᶻatuɬ əlgʷəʔ čaʔkʷ, tiʔiɬ ʔalalš

ƛ’u=pədᶻatuɬ əlgʷəʔ čaʔkʷ tiʔiɬ ʔal–alš–s

HAB=duck.hunt PL seaward DIST DSTR–cross.sex.sibling–3PO

‘they would be out duck-hunting on the sea, her brothers’

(Hess, 2006: 17, line 131)

b. gʷəl ləčadᶻil t’aq’təxʷ

gʷəl lə=čadᶻ–il t’aq’t=əxʷ

SCONJ PROG=be.hidden–INCH landward=now

‘and they are hiding up inland’

(Hess, 2006: 54, line 288)

c. saq’ʷ dxʷšəq tiʔiɬ c’ix%c’ix%

saq’ʷ dxʷ–šəq tiʔiɬ c’ix%c’ix%

fly CNTRPT–high DIST fish.hawk

‘Fish Hawk flies up high’

(Hess, 1995: 151, line 11)

d. ʔuʔux%ʷtxʷ čəd dxʷlil

ʔu–ʔux%ʷ–txʷ čəd dxʷ–lil

PFV–go–ECS 1SG.SUB CNTRPT–far

‘I took it a ways off’

(Bates, 1994: 138)

Occasionally, clauses may contain more than one locative adverb:

(258) ʔəsɬaɬlil lil čaʔkʷ ʔal tiʔəʔ hikʷ x%ʷəlč

ʔəs–ɬaɬlil lil čaʔkʷ ʔ al tiʔəʔ hikʷ x%ʷəlč

STAT–live far seaward PR PROX big sea

’she dwelled way out in the great sea’

(Bierwert 1996: 183, line 44)

Although locative adverbs follow the verb that they modify, their ordering with respect to other

elements in the clause is flexible. Compare the position of the adverb relative to the subject NP

in (257c) and (259a), and the position of the adverb relative to the plural marker əlgʷəʔ in (257a)

and (259b):

232

(259) a. tuləʔibəš tiʔiɬ bəščəb liɬʔilgʷiɬ

tu=lə=ʔibəš tiʔiɬ bəščəb liɬ–ʔilgʷiɬ

PAST=PROG=travel DIST mink PRLTV–ashore

‘Mink was traveling along the shore’

(Hess 1995: 82)

b. tuʔux%ʷtub dxʷšəq əlgʷəʔ

tu=ʔux%ʷ–txʷ–b dxʷ–šəq əlgʷəʔ

PAST=go–ECS–PASS CNTRPT–high PL

‘they were taken up’

[DS Star Child, line 20]

In addition, one of the locative adverbs is attested in some examples in sentence-initial position:

(260) a. lil čəxʷ ʔiɬqʷiq’ʷ tul’ʔal ʔəca

lil čəxʷ ʔiɬ–qʷiq’ʷ tul’–ʔal ʔəca

far 2SG.SUB PRTV–strong CNTRFG–at I

‘you are stronger than I’

b. lil čəxʷ ʔiɬsadᶻəp tul’ʔal ʔəca

lil čəxʷ ʔiɬ–s–hadᶻ•əp tul’–ʔal ʔəca

far 2SG.SUB PRTV–NP–long•bottom CNTRFG–at I

‘you are taller than I’

(Bates, 1994: 137)

Both of these sentences, however, are comparative constructions and the adverb lil does not have

its literal meaning of ‘far’ (i.e., ‘greater distance’), but instead has a more metaphorical,

intensifying meaning (‘greater in quality X’). Outside of this type of fixed expression, lil patterns

with the other locative adverbs.

Like lexical adverbs, locative adverbs are also potential clausal predicates:

(261) a. čaʔkʷ kʷi šuƛ’

čaʔkʷ kʷi šuƛ’

seaward REM ebb.tide

‘the tide is out’

[ML Mink and Tutyika I, line 122]

b. dxʷʔaɬx%ad kʷədiʔ sʔux%ʷs

dxʷ–ʔaɬx%ad kʷədiʔ s=ʔux%ʷ=s

CNTRPT–downstream REM.DMA NM=go=3PO

‘he moves downstream’ (lit. ‘his going yonder [is] towards downstream’)

(Hess 2006: 36, line 364)

233

c. tul’q’ixʷ tiʔiɬ saʔsaʔ

tul’–q’ixʷ tiʔiɬ saʔ–saʔ

CNTRFG–upstream DIST DSTR–bad

‘it is bad upstream’ (lit. ‘that which [are] bad [are] from upstream’)

[ML Basket Ogress, line 208]

Locative adverbs are also found in predicate position in subordinate clauses:

(262) a. ɬusʔəɬəd čəxʷ ʔə kʷi č’itəxʷ ʔaciɬtalbixʷ

ɬu=sʔəɬəd čəxʷ ʔə kʷi č’it=əxʷ ʔaciɬtalbixʷ

IRR=food 2SG.SUB PR REM nearby=now people

‘you will be food for the people who are nearby’

[ML Mink and Tutyika I, line 246]

b. ʔa kʷi sʔiɬčaʔkʷs

ʔa kʷi s=ʔiɬ–čaʔkʷ=s

be.there REM NM=PRTV–seaward=3PO

‘and there he was down by the water’

(Hess 1998: 65, line 17)

c. ʔəx%idəxʷ tə xʷiʔ ləhaʔɬ ʔudxʷluq’ʷucutigʷəd kʷi dəxʷč’its kʷi bəsucucuts

ʔəx%id=əxʷ tə xʷiʔ lə=haʔɬ ʔu–dxʷ–luq’ʷucut•igʷəd kʷi

what.happen=now NSPEC NEG NEGP=good PFV–CTD–make.sound•body REM

dəxʷ=č’it=s kʷi bə=s=ʔu–cut–cut=s

ADNM=nearby=3PO REM ADD=NM=PFV–DSTR–say=3PO

‘why is that no-good one making noises in his body [as] he draws near speaking over

and over?’

[ML Mink and Tutyika II, line 32]

d. šušɬbitəbəxʷ tiʔiɬ ɬuʔalx%adəs

šu–šɬ–bi–t–əb=əxʷ tiʔiɬ ɬu=ʔalx%ad=əs

ATTN–see–MAP–ICS–PASS=now DIST IRR=downstream=3SBJ

‘he is watched for as he comes downstream’

[MW Star Child, line 80]

As shown in these examples, locative adverbs can be the predicates of relative clauses (262a), of

both types of nominalized clauses (262b) and (c), and of subordinate subjunctive clauses (262d).

Another property that locative adverbs share with lexical adverbs is the potential to take

verbal derivational affixes; however, several of the locative adverbs seem to do so much more

prolifically, having a derivational potential much more akin to that of verbal radicals:

234

√čaʔkʷ ‘seaward’

čagʷəb ‘be at sea’

čagʷəd ‘take  out to sea’

čagʷil ‘get out to sea’

čaʔkʷdxʷ ‘manage to get  to sea’

čaʔkʷtxʷ ‘take  out to sea’

dxʷčaʔkʷtxʷ ‘take  seaward’

√č’it ‘nearby’

č’itcut ‘come close, approach’

č’itil ‘draw near’

č’itis ‘approach ’

√lil ‘far’

lild ‘move away’

lildxʷ ‘draw away from ’

lis ‘go over to ’

√šq ‘high’

šəqəd ‘move  up high’

šəqlax&adəb ‘raise arms’

šqil ‘rise up’

√gʷəd ‘down’

gʷədil ‘sit down’

gʷədiltxʷ ‘sit  down’

gʷədiluɬ ‘go there to sit down’

gʷədis ‘sit down next to ’

√t’aq’t ‘inland, landward’

dxʷt’aq’tcut ‘take oneself to higher ground’

dxʷt’aq’txʷ ‘take  ashore’

Table 63: Verb stems derived from locative adverbs

Also like verbs, locative adverbs can take aspectual inflection when used as predicates:

(263) a. gʷəl ʔučaʔkʷ tsi sxʷiyuk’ʷ

gʷəl ʔu–čaʔkʷ tsi sxʷiyuk’ʷ

SCONJ PFV–seaward SPEC:FEM Basket.Ogress

‘and the Basket Ogress came down to shore [from inland]’

[JS Basket Ogress, line 31]

b. lədxʷq’ixʷ čəd

lə=dxʷ–q’ixʷ čəd

PROG=CNTRPT–upstream 1SG.SUB

‘I’m going upstream’

(Bates, 1994: 187)

The most frequent aspect-marker found with locative adverbial predicates is the progressive,

although — as shown in (263a) — others are possible.

Another characteristic of the locative adverbs that sets them apart from verbs and from other

adverbs is their appearance as the bare complements of PPs such as those in (264):

(264) a. ʔup’ayəq ʔal kʷədiʔ t’aq’t

ʔu–p’ayəq ʔal kʷədiʔ t’aq’t

PFV–carve.canoe at REM.DMA inland

‘he was carving a canoe up on shore’

[ML Mink and Tutyika II, line 62]

235

b. tuɬiq’ʷtəb k’ʷəɬ ʔə tul’ʔaɬx%ad kʷi tudsqa

tu=ɬiq’ʷ–t–əb k’ʷəɬ ʔə tul’–ʔaɬx%ad kʷi

PAST=hooked–ICS–PASS PTCL PR CNTRFG–downstream REM

tu=d–sqa

PAST=1SG.PO–older.sibling

‘my older brother was kidnapped [and taken] downstream’

[DS Star Child, line 210]

c. ƛ’ubəxʷ čəxʷ x%ʷul’əxʷ ɬubəščəb, ɬup’aƛ’aƛ’, ɬusdukʷ ɬuʔal tudiʔ čaʔkʷ

ƛ’ub=əxʷ čəxʷ x%ʷul’=əxʷ ɬu=bəščəb ɬu=p’aƛ’aƛ’ ɬu=sdukʷ

well=now 2SG.SUB just=now IRR=mink IRR=worthless IRR=riff.raff

ɬu=ʔal tudiʔ čaʔkʷ

IRR–PR DIST.DMA seaward

‘you should be just a mink, a no-account, riff-raff down there by the water’

(Hess 1998: 69, line 122)

This sort of distribution is unusual in that, like verbs, other types of adverbs in this position are

either nominalized or have readings as headless subject-centred relative clauses. A reading as a

headless relative clause is precisely what is found when locative adverbs head embedded clauses

acting as arguments. Compare (264c) and the sentence in (265):

(265) ʔəxʷčəgʷasəb k’ʷəɬ tiʔəʔ ʔučaʔkʷ

ʔəs–dxʷ–čəgʷas–əb k’ʷəɬ tiʔəʔ ʔu–čaʔkʷ

STAT–CTD–wife–DSD PTCL PROX PFV–seaward

‘this one who came down to the coast, it seems, wants a wife’

(Hess 1998: 97, line 180)

Thus, it seems that the retention of the adverbial reading for locative adverbs used as

complements is particular to PPs and does not apply to the argument position of verbs.

When used as the complements of prepositions, verbs and adverbs also require either a

determiner or are introduced by a demonstrative adverbial. Locative adverbs in PPs are more

often than not introduced by a demonstrative adverbial as well (cf. 264a and c), but can also

appear on their own (264b). Only one of the locative adverbials, ʔilgʷiɬ ‘ashore’, is occasionally

attested with an adverbial reading as the complement of a preposition introduced by a

determiner:

236

(266) qaha ləcuʔukʷukʷ ʔal tiʔəʔ ʔilgʷiɬ ʔə tiʔəʔ stuləkʷ

qa–ha ləcu–ʔukʷukʷ ʔal tiʔəʔ ʔilgʷiɬ ʔə tiʔəʔ stuləkʷ

INTNS–many CONT–play at PROX ashore PR PROX river

‘many were playing by the shore of the river’

[DM Basket Ogress, line 10]

In these cases, however, the interpretation of ʔilgʷiɬ is actually that of a noun, ‘shore’, rather than

a locative adverb and examples like (266) probably represent the elision of the nominalizing

prefix s- from the form sʔilgʷiɬ ‘shore’, making the phrase tiʔəʔ ʔilgʷiɬ ʔə tiʔəʔ stuləkʷ ‘the

shore of the river’ an ordinary NP.

Although the locative adverbs are essentially static in meaning, designating a region in which

their modificand is located, they are very frequently found compounded with the directional

particles (Section 2.7.2), dxʷ ‘centripetal’, tul’ ‘centrifugal’, and liɬ ‘prolative’ to form more

dynamic expressions:

(267) a. ɬik’ʷitəb dxʷʔaɬx%ad

ɬik’ʷi–t–əb dxʷ–ʔaɬx%ad

hooked–ICS–PASS CNTRPT–downstream

‘he was taken by force downstream’

[DS Star Child, line 133]

b. tuɬik’ʷtəb ʔə kʷi tul’ʔaɬx%ad sɬəɬadəyʔ

tu=ɬik’ʷ–t–əb ʔə kʷi tul’–ʔaɬx%ad sɬə–ɬadəyʔ

PAST=hooked–ICS–PASS PR REM CNTRFG–downstream PL–woman

‘he was taken by force by some women from downstream’

[DS Star Child, line 203]

c. gʷəl ƛ’ubəliɬčaʔkʷ kʷi səʔibəš liɬʔilgʷiɬ

gʷəl ƛ’u=bə=liɬ–čaʔkʷ kʷi s=lə=ʔibəš=s liɬ–ʔilgʷiɬ

SCONJ HAB=ADD=PRLTV–seaward REM

NM=PROG=travel=3PO PRLTV–ashore

‘and he would be down by the sea again traveling along the shore’

(Hess 1998: 67, line 55)

The effect of combining the directional particle and the locative adverb is one of creating an

adverbial expression of atelic motion in a particular direction or (in the case of the prolative)

through a region. Several of these compound locative adverbs can be grouped into near-

synonymous pairs which in turn correspond to certain verbs of motion:

dxʷčaʔkʷ ‘towards seaward’

tul’t’aq’t ‘from inland’

k’ʷit ‘go down to the sea’

237

dxʷt’aq’t ‘towards inland’

tul’čaʔkʷ ‘from seaward’

čubə ‘go inland’

dxʷʔaɬx&ad ‘towards downstream’

tul’q’ix ‘from upstream’

qʷic ‘go downstream’

dxʷq’ix ‘towards upstream’

tul’ʔaɬx&ad ‘from downstream’

təyil ‘go upstream’

Table 64: Compound locative adverbs and related motion verbs

Within these triads, the verbs are typically more specific than the adverbs in that they imply

either a point of origin and/or a specific type of travel. For the two sets of marine terms, the

verbs k’ʷit ‘go down to the sea’ and čubə ‘go inland’ contrast with the adverbs in that, in

addition to direction of motion, they specify either an endpoint or a point of origin as well as the

notion of traveling on land, whereas the adverbs are more general. Consider the diagram in

Figure 3:

Figure 3: Expressions for marine orientation

As shown here, the verb k’ʷit expresses motion beginning at some indefinite point inland and

terminating at the shoreline, whereas the adverbs dxʷčaʔkʷ and tul’t’aq’t apply to any movement

in a direction from away from inland towards the sea or out to sea, irrespective of where that

movement begins or ends. Similarly, čubə expresses atelic motion beginning at the shoreline,

while the adverbs dxʷt’aq’t and tul’čaʔkʷ are applicable to any movement away from the sea

directed towards land or farther inland. The static terms t’aq’t and čaʔkʷ express the two cardinal

directions ‘inland’ and ‘seaward’, respectively, and ʔilgʷiɬ is used as a locative expression

indicating location on or near the seashore.

In the case of the two sets of riverine terms, the situation is slightly different. In these cases,

the verbs are also more specific than the adverbs; however, with the riverine terms, the verbs do

dxʷt’aq’t

tul’čaʔkʷ

dxʷčaʔkʷ

tul’t’aq’t

čaʔkʷ

t’aq’t

k’ʷit

čubə

ʔilgʷiɬ

238

not specify a point of origin, but only that the travel took place on the river. The adverbs, on the

other hand, specify only the direction of travel, which could have taken place on the river or

along the shore. This is illustrated in Figure 4:

Figure 4: Expressions for riverine orientation

It should be noted that the term ʔilgʷiɬ ‘shore’ is equally applicable to the sea shore and to the

river bank (see, for example, the sentence in 266 above).

2.5.3 Predicate particles

Like many other Northwest Coast languages, Lushootseed has a number of elements that

appear obligatorily in sentence-second position. While some of these (like the subjunctive

subject markers — Section 8.1.1), are bound enclitics that become phonologically dependent on

the first word of the clause in which they appear, others are phonologically more independent

elements which occur as the second word in a clause.

These particles include matrix-clause

subject markers (see Table 79 in Section 8.1.1 below) and a small class of predicate particles:

ʔu ‘interrogative’

dᶻəɬ ‘seemingly’

əw’ə ‘so!’ (mirative)

kʷaʔ ‘probably’

k’ʷəɬ ‘it is said’ (quotative)

sixʷ ‘again, as usual’ (exasperation)

In strictly phonological terms, these particles are also clitics in the sense of being unstressed, phonologically

dependent elements; however, unlike enclitics, sentence-second particles can attach themselves either to the

preceding or to the following word, depending on prosodic context and other factors. See Beck (1999) for further

discussion.

ʔaɬx%ad

q’ixʷ

təyil

dxʷq’ixʷ

tul’ʔaɬx&ad

qʷic

dxʷʔaɬx%ad

tul’q’ixʷ

ʔilgʷiɬ

239

uʔxʷ ‘still, yet’

Table 65: Predicate particles

As sentence-second elements, predicate particles follow the sentence predicate when this is the

first element in a clause; if the predicate is modified by an adverb or some other element or

elements, the predicate particle generally appears immediately following the first of these

modifying elements:

(268) a. ʔəsx%əɬ dᶻəɬ tsiʔəʔ k’aʔk’aʔ

ʔəs–x%əɬ dᶻəɬ tsiʔəʔ k’aʔk’aʔ

STAT–sick PTCL PROX:FEM crow

‘Crow must be sick’

(Hess 1995: 88, ex. 3)

b. yaw’ dᶻəɬ bəliɬč’it kʷi sədxʷluq’ʷucutigʷəds čxʷəluʔ

yaw’ dᶻəɬ bə=liɬ–č’it kʷi s=lə=dxʷluq’ʷucutigʷəd=s čxʷəluʔ

until PTCL ADD=PRLV–be.near REM NM=PROG=sound.in.body=3PO whale

‘as he drew ever closer Whale must have been making sounds in his body’

[ML Mink and Tutyika I, line 24]

In both of the sentences in (268), the predicate particle dᶻəɬ ‘seemingly’ appears in sentence-

second position, in (268a) following the verbal sentence predicate and in (268b) following a

sentence-initial adverb, yaw’ ‘until’. When there is more than one adverb in a sentence, the most

frequent pattern is for the predicate particles to follow the first adverb:

(269) a. day’əxʷ dᶻəɬ higʷəxʷ ʔuk’ʷəɬk’ʷɬatəb ti tudqʷuʔ

day’=əxʷ dᶻəɬ higʷ=əxʷ ʔu–k’ʷəɬ–k’ʷɬa–t–əb ti tu=d–qʷuʔ

only=now PTCL big=now PFV–DSTR–spilled–ICS–PASS SPEC PAST=1SG.PO–water

‘indeed, he must have spilled all my water’

(Hess 1995: 90, ex. 23)

b. ƛ’ub čəɬ ʔu x%ʷul’ ɬut’uk’ʷ

ƛ’ub čəɬ ʔu x%ʷul’ ɬu=t’uk’ʷ

well 1PL.SUB INT only IRR=go.home

‘should we just go home?’

(Hess 1995: 90, ex. 22)

240

Although the order in (269) is the most common, it is also possible for the predicate particle to

be placed following a non-initial adverb, as are the particles əw’ə ‘so!’ in (270a) and sixʷ ‘again,

as usual’ in (270b):

(270) a. cick’ʷ k’ʷəɬ xʷuʔələʔ həlaʔb əw’ə ʔəsx%əɬ tsiʔəʔ k’aʔk’aʔ

cick’ʷ k’ʷəɬ xʷuʔələʔ həlaʔb əw’ə ʔəs–x%əɬ tsiʔəʔ k’aʔk’aʔ

again QTV supposedly extremely PTCL STAT–sick PROX:FEM crow

‘so I guess they say that Crow is really very sick’

(Hess 1995: 91, ex. 25)

b. cick’ʷ xʷuʔələʔ sixʷ həlaʔb ʔəstagʷəxʷ tiʔəʔ qaw’qs

cick’ʷ xʷuʔələʔ sixʷ həlaʔb ʔəs–tagʷ=əxʷ tiʔəʔ qaw’qs

again maybe PTCL extremely STAT–hungry=now PROX raven

‘I guess Raven is really very hungry again’

(Hess 1995: 91, ex. 26)

In general terms, the placement of the particles in sentences like those in (270) seems to be

related to the meaning of the sentence, in particular the semantic scope of the predicate particle:

specifically, when the particle is primarily a modifier of the sentence predicate, it occurs in

sentence-second position, and when it is a modifier of a particular adverb it follows that element.

The distinctions between different placements of predicate particles, however, are extremely

subtle and difficult to unravel based on English translations, even when using contextualized

examples from texts, and so any broader generalizations that might be made on the issue would

be largely speculative.

It is also possible for more than one predicate particle to appear in sequence in the same

clause:

(271) a. ʔəsx%əɬ uʔxʷ čəxʷ ʔu

ʔəs–x%əɬ uʔxʷ čəxʷ ʔu

STAT–sick PTCL 2SG.SUB INT

‘are you still sick?’

See also the sentence in (246) above, where the matrix subject-marker follows the second, rather than the first,

adverb in the sentence (contra the usual case, in which the subject-marker follows the first adverb, sticking to strict

sentence-second position). On the whole, however, the matrix subject-markers and the interrogative particle are

much more strictly sentence-second than the other predicate particles and show significantly less variation of the

type illustrated in (269).

241

b. ʔəshəliʔtub uʔxʷ dᶻəɬ čəxʷ

ʔəs–həliʔ–txʷ–b uʔxʷ dᶻəɬ čəxʷ

STAT–alive–ECS–PASS PTCL PTCL 2SG.SUB

‘you must still be kept alive’

(Hess 1995: 88, ex. 7–8)

c. gʷəɬ yaw’ čəxʷ k’ʷəɬ ʔupigʷəd čxʷa ɬuhəliʔ

gʷəɬ yaw’ čəxʷ k’ʷəɬ ʔu–pigʷəd čxʷa ɬu=həliʔ

SCONJ if.only 2SG.SUB PTCL PFV–sing.power.song 2SG.COORD IRR=alive

‘and they say only if you put on a spirit-power ceremony will you recover’

(Hess 1998: 58, line 47)

d. ʔəsx%əɬ k’ʷəɬ əw’ə dᶻəɬ ʔu sixʷ tsi k’aʔk’aʔ

ʔəs–x%əɬ k’ʷəɬ əw’ə dᶻəɬ ʔu sixʷ tsi k’aʔk’aʔ

STAT–sick PTCL PTCL PTCL INT PTCL SPEC:FEM crow

‘so does it seem that Crow is sick as usual as they say?’

(Hess 1995: 88, ex. 10)

In terms of the relative ordering of predicate particles, there is a great deal of permissible

variation, although matrix-subject markers always precede the interrogative ʔu when these co-

occur, and the particle uʔxʷ ‘still’ always precedes either or both of these, as in (271a). Both the

matrix subject-markers and the interrogative particle are variably ordered with respect to the

other predicate particles (cf. 271a versus b and c). Otherwise, the quotative particle k’ʷəɬ tends to

be the first in any sequence of particles and sixʷ ‘again, as usual’ tends to be last (271d);

however, these are only tendencies and the relative ordering of the particles can be manipulated

for emphatic and stylistic effect — as, for example, in (272), where sixʷ is the first particle in the

string:

(272) ʔəsx%əɬ sixʷ dᶻəɬ əw’ə tsiʔəʔ k’aʔk’aʔ

ʔəs–x%əɬ sixʷ dᶻəɬ əw’ə tsiʔəʔ k’aʔk’aʔ

STAT–sick PTCL PTCL PTCL PROX:FEM crow

‘it seems that Crow must be sick, again’

(Hess 1995: 88, ex. 9)

The rhetorical effect of this sentence, which drips sarcasm, is largely achieved by placing sixʷ

immediately after the verb, ahead of all the other particles. As with the placement of predicate

particles relative to adverbs, the nuances expressed by the different orderings of predicate

particles are unlikely to be easily accessible through English glosses of sentences. Further insight

242

into the issue, if at all possible, will have to await more detailed stylistic and rhetorical analysis

of texts and other forms of recorded discourse.

2.5.4 Demonstrative adverbials

Demonstrative adverbials form a class of words that have some of the properties of both

demonstrative determiners (2.4) and lexical adverbs (Section 2.5.1). The complete set of these

elements listed in the Lushootseed Dictionary is given in Table 66:

PROX

DIST

REM

PROX:UNQ

DIST:UNQ

MASC

diʔaʔ

dišaʔ (SL)

tudiʔ

tadiʔ

kʷədiʔ

tiʔaʔ*

taʔa

FEM

—

tsudiʔ

tsadiʔ*

—

tsiʔaʔ*

tsaʔa*

*unattested in the present corpus

Table 66: Demonstrative adverbials

As can be seen in Table 66, demonstrative adverbials coincide with the specific determiners

(Table 58) in terms of the semantic distinctions they express. Like the determiners, they encode

three purely spatial categories — proximal, distal, and remote — and (in three cases) they make

distinctions in natural gender between non-feminine/unmarked and feminine when used in lieu of

determiners to introduce referential expressions (see 275 below) or as independent pronominals

(278).

They also seem to mark the category of uniqueness encoded by the determiner tiʔacəc,

although unlike the determiners, the demonstrative adverbials combine this category with two

degrees of spatial deixis. Unfortunately, of the four unique forms, only one — the distal unique

taʔa — is attested in the present corpus: neither its feminine form nor either of the contrasting

proximal forms, tiʔaʔ and tsiʔaʔ, has been found in any of the analyzed texts, although examples

containing tiʔaʔ are found in Hess & Hilbert (1976) (see example 678 in Section 8.8 below). Of

the remaining forms in Table 66, the distal feminine tsadiʔ is unattested, tsudiʔ is attested once,

Presumably, the feminine gender also applies in diminutive and affective contexts, as it does with deictic

determiners, although there are no attestations of this usage in the corpus.

243

the proximal diʔaʔ and the distal tadiʔ have only four attestations each, and taʔa has six

attestations.

Only the remote kʷədiʔ and the distal form tudiʔ are at all frequent in the corpus.

This makes it somewhat difficult to draw robust generalizations about the meanings and uses of

the demonstrative adverbials, and suggests that the class of demonstrative adverbials is

disintegrating as some of its members drop out of use and others acquire functions more typical

of other lexical classes.

The basic, most consistent role of the demonstrative adverbials is that of lexical adverb, as

shown in (273):

(273) a. tsudiʔ ʔəƛ’a(hə)xʷ

tsudiʔ ʔəƛ’a=həxʷ

DIST.DMA:FEM come.to=now

‘here she comes’

[DS Star Child, line 288]

b. ƛ’asčal diʔaʔəxʷ kʷaʔ

ƛ’u=ʔas–čal diʔaʔ=əxʷ kʷaʔ

HAB=STAT–how PROX.DMA=now PTCL

‘they have been here, though’ (lit. ‘[they] have been here in some manner’)

(Hess 2006: 44, line 63)

c. q’ʷəlilcəxʷ tudiʔ diʔi tə wiw’su

q’ʷəl•ilc–c=əxʷ tudiʔ diʔi tə wiw’su

warm•round–ALTV=now DIST.DMA far NSPEC children

‘the stones are warm there now for the children’

[ML Basket Ogress, line 58]

d. x%ʷul’əxʷ čəxʷ ʔəsxʷəɬ sƛ’alabac kʷ(i) adsʔux%ʷ, tiʔiɬ ƛ’usǰuʔadadəxʷ kʷədiʔ diʔiʔ tiʔiɬ

ƛ’(u)adsəsɬidɬidšəd

x%ʷul’=əxʷ čəxʷ ʔəs–xʷəɬ s=ƛ’al•abac kʷi ad=s=ʔux%ʷ tiʔiɬ

only=now 2SG.SUB STAT–lack NM=cover•body REM 2SG.PO=NP–go DIST

ƛ’u=sǰuʔadad=əxʷ kʷədiʔ diʔiʔ tiʔiɬ ƛ’u=ad=s=ʔəs–ɬid–ɬid•šəd

HAB=regalia=now REM.DMA far DIST HAB=2SG.PO=NM=STAT–DSTR–tie•leg

‘just go without clothes, the ceremonial accessories there that you’ve tied to your legs’

(Hess 2006: 26, line 82)

All six of these are transcribed in the original sources as ta, a particle meaning (roughly) ‘just this one’; however,

the environments in which they appear are not those expected for a predicate particle and on closer listening the

form on tape is heard to be [taʔa] in the five of six cases where the recording is available in accessible format. The

sixth form, given in (277c), is judged also to be taʔa based on meaning and syntactic environment

244

In each of the examples here, a demonstrative adverbial appears as a modifier of a verbal

predicate, specifying a spatial location for an action or a state relative to the speaker or speech

act. In (273a), the demonstrative adverb tsudiʔ precedes the verb, as expected for an adverb

(Section 2.5.1); however, in the bulk of their attestations, demonstrative adverbials follow the

predicate they modify, as is more typical of locative adverbs (2.5.2). All of the demonstrative

adverbials, with the exceptions of taʔa and tadiʔ, are found in this role in the texts in the current

corpus; however, the most frequent demonstrative adverb overall, kʷədiʔ, is only clearly attested

in this role in one sentence, given in (273d).

Demonstrative adverbials are also potentially clausal predicates:

(274) a. ɬudiʔaʔəxʷ kʷi qa ʔaciɬtalbixʷ

ɬu=diʔaʔ=əxʷ kʷi qa ʔaciɬtalbixʷ

IRR=PROX.DMA=now REM many person

‘there will be a lot of people here’

(Hess 2006: 18, line 172)

b. huy čələp ʔuhəliʔdubuɬəxʷ ʔə tiʔəʔ sdiʔaʔləp

huy čələp ʔu–həliʔ–dxʷ–buɬ=əxʷ ʔə tiʔəʔ s=diʔaʔ=ləp

SCONJ 2PL.SUB PFV–alive–DC–1PL.OBJ=now PR PROX NM=PROX.DMA=2PL.PO

‘so you folks gave us life by your being here’

(Hess 2006: 78, line 884)

c. tadiʔəxʷ ti bəsuʔux%ʷs

tadiʔ=əxʷ ti bə=s=ʔu–ʔux%ʷ=s

DIST.DMA=now SPEC ADD=NM=PFV–go=3PO

‘it has gone far now’

(Hess 2006: 26, line 99)

All of the attested forms from Table 66 are found in this role with the exception of kʷədiʔ and

taʔa. The paucity of attestations of the latter make this gap in the data unremarkable,

although

The fact that taʔa, the only attested unique demonstrative adverbial, does not appear in the role of either adverb or

predicate may be an indication that these four words belong to a separate lexical class, perhaps representing a subset

of the deictic determiners. However, as noted above, only one of the four appears in the corpus at all and that one is

poorly attested, so the absence of attestations of these in particular syntactic roles may only be an accidental gap in

the data. These words are grouped with the demonstrative adverbials based on morphological considerations and

Thom Hess’s intuitions about their behaviour drawn from elicitation, but may need to be amended in the light of

future data.

245

the absence of the former in a predicative role, given its relative frequency, is a little surprising

and may be related to its infrequent use as an adverb.

The second environment in which demonstrative adverbials is found is that of determiner.

When filling this role, the adverbial takes the place of an ordinary determiner (Section 2.4) and

directly introduces an NP or other type of referential expression. This type of construction is by

far the most common in PPs:

(275) a. ɬalisəbš, ti dsbalucid, čəxʷə tuliltubš dxʷʔal tudiʔ wiw’suʔ

ɬalil–s–əbš ti d–sbalucid čəxʷə

go.ashore–ALTV–1SG.OBJ SPEC 1SG.PO–in.law 2SG.COORD

tulil–txʷ–bš dxʷʔal tudiʔ wiw’suʔ

cross.river–ECS–1SG.OBJ CNTRPT–at DIST.DMA children

‘take me ashore, my in-law, and take me across the river to those children’

[JS Basket Ogress, line 33]

b. huy bəʔux%ʷ tiʔiɬ ƛ’uciqʷ əlgʷəʔ ʔə kʷədiʔ stab

huy bə=ʔux%ʷ tiʔiɬ ƛ’u=ciqʷ əlgʷəʔ ʔə kʷədiʔ stab

SCONJ ADD=go PROX HAB=dig.roots PL PR REM.DMA what

‘then they again went to root-dig for those things’

(Hess 2006: 38, line 405)

c. xʷiʔ gʷəsutəɬəɬs dxʷʔal taʔa ʔaciɬtalbixʷ

xʷiʔ gʷə=s=ʔu–təɬəɬ=s dxʷ–ʔal taʔa ʔaciɬtalbixʷ

NEG SBJ=NM=PFV–arrive.safely=3PO CNTRPT–at DIST.DMA:UNQ person

‘it could not arrive safely to those very people [for whom it was intended]’

(Hess 1998: 63, line 81)

Whereas kʷədiʔ is poorly attested in the adverb-like roles of demonstrative adverbials, its most

frequent attestation is in constructions such as those shown in (275), where it shows considerable

overlap with the remote demonstrative determiner kʷi. Also well-attested in this role is tudiʔ, and

all but two of the six attestations of taʔa function as determiners introducing the complements of

This sentence appears in the source as xʷiʔ gʷəsutəɬəɬs ʔə tiʔiɬ sʔələd dxʷʔal ta ʔaciɬtalbixʷ. The PP ʔə tiʔiɬ

sʔələd ‘of the food’ was added post-recording by the narrator’s husband while the story was being transcribed and is

an aid to making the sentence clearer. The change from ta to taʔa reflects a correction to the transcription (see

footnote 88 above).

246

PPs. Less frequently, some of the demonstrative adverbials serve as determiners for ordinary NP

arguments of verbs, as in (276):

(276) a. til’əbəxʷ ʔu ʔəca, cən’əx%imaligʷəd, gʷəl xʷiʔ kʷi gʷəsuʔəy’dxʷs kʷədiʔ

sʔumʔumən’iʔ

til’əb=əxʷ ʔu ʔəca cən’əx%imaligʷəd gʷəl xʷiʔ kʷi

immediately=now INT I Cən’əx%imaligʷəd SCONJ NEG REM

gʷə=s=ʔu–ʔəy’dxʷ=s kʷədiʔ sʔub–ʔubadiʔ

SBJ=NM=PFV–find=3PO REM.DMA DSTR–hunter

‘Am I, Cən’əx%imaligʷəd, not the one to find the hunters right away?’

(Hess 1998: 84, lines 172 – 173)

b. tudiʔ gʷəd, diɬ dəxʷtul’ʔa čəɬ tudiʔ gʷəd swatixʷtəd

tudiʔ gʷəd diɬ dəxʷ=tul’–ʔa čəɬ tudiʔ

DIST.DMA downward FOC ADNM=CNTRFG–be.there 1PL.SUB DIST.DMA

gʷəd swatixʷtəd

downward land

‘way down, that is where we are from, that world way down [there]’

[DS Star Child, line 37]

Only kʷədiʔ and tudiʔ are attested in this environment, although the others are so infrequent that

this may not be a significant gap in the data. kʷədiʔ, tudiʔ and taʔa are also found introducing

headless relative clauses:

(277) a. hikʷ q’il’bid tudiʔ ləʔəƛ’

hikʷ q’il’bid tudiʔ lə=ʔəƛ’

big canoe DIST.DMA PRG–come

‘what is coming over there [is a] big canoe’

(Hess 2006: 54, line 297)

b. qʷiʔad kʷədiʔ diʔucid

qʷiʔad kʷədiʔ diʔ•ucid

call.out REM.DMA opposite.side•river

‘the one across the river called out’

(Hess 1998: 96, line 148)

The pronunciation of sʔubʔubadiʔ ‘hunters’ in this line reflects Raven’s speech.

247

c. bəščəb taʔa tuləʔibəš

bəščəb taʔa tu=lə=ʔibəš

mink DIST.DMA:UNQ PAST=PROG=travel

‘that very one traveling was Mink’

(Hess 1998: 65, line 2)

tudiʔ and taʔa are found in the corpus used as pronominal complements to prepositions, and

tiʔaʔ turns up in this role in decontextualized examples in Hess & Hilbert (1976):

(278) a. ʔahaʔ kʷi adsʔa ʔal tudiʔ

ʔahaʔ kʷi ad=s=ʔa ʔal tudiʔ

right.there REM 2SG.PO=NM=be.there at DIST.DMA

‘there is your place over there’

(Hess 2006: 34, line 313)

b. gʷəl ʔux%ʷtub tiʔəʔ bədaʔs ʔə tudiʔ

gʷəl ʔux%ʷ–txʷ–b tiʔəʔ bədaʔ–s ʔə tudiʔ

SCONJ go–ECS–PASS PROX child–3PO PR DIST.DMA

‘so his son is taken way [up]’

(Hess 2006: 26, line 98)

c. dᶻubud dəxʷʔa ʔə taʔa

dᶻubud dəxʷ=ʔa ʔə taʔa

kick–ICS ADNM=be.there PR DIST.DMA:UNQ

‘she kicked it where it was right there’

[DS Star Child, line 82]

d. ʔiɬx%əb tiʔəʔ wəq’əb ʔal tiʔaʔ dxʷʔal tiʔiɬ wəq’əb ʔal tudiʔ

ʔiɬ–x%əb tiʔəʔ wəq’əb ʔal tiʔaʔ dxʷ–ʔal tiʔiɬ wəq’əb

PRTV–heavy PROX box at PROX:UNQ.DMA CNTRPT–at DIST box

ʔal tudiʔ

at DIST.DMA

‘this chest right here is heavier than that chest over there’

(Hess & Hilbert 1976: II, 118)

Thus, for tudiʔ there is almost complete overlap in the distribution with the determiners, while

kʷədiʔ is most frequently attested acting as a determiner in PPs and NPs, and taʔa is found

serving as a pronominal and as a determiner in PPs and introducing a headless relative clause. Of

The second word in this example is transcribed as [ta] in the cited source, although [taʔa] is in fact (barely)

audible on the tape (see footnote 88 above).

The original transcription of the last word in this line is [ta]; however, the particle ta ‘just this one’ would not be

grammatical in this context, whereas the demonstrative adverbial taʔa would be (see footnote 88 above). The

recording is not currently available in accessible format and so can not be checked at this time.

248

course, taʔa is very poorly-attested in the current corpus, as are several of the other

demonstrative adverbial forms, and the absence of examples in one or the other of the syntactic

roles attested for the other demonstrative adverbials may simply be an accidental gap in the data.

More robust generalizations about their behaviour will have to await the discovery of further

attestations in as yet unanalyzed texts.

2.6 Interrogative/indefinite words

Lushootseed has a set of words, drawn from a variety of lexical classes, that are used both as

indefinite anaphoric expressions (‘someone’, ‘something’, ‘somewhere’, etc.) and as

interrogative words in information questions (Section 8.4.2). In addition to having the syntactic

distribution of whichever lexical class they belong to, these interrogative words also take on

some of the morphosyntactic properties of verbs when used as the syntactic predicates of

information questions (Section 8.4.2). In total, there are 17 of these, given in Table 67:

ʔəx&id ‘ what happened?; something happens;’

dxʷčad ‘to where?; to somewhere’

ʔəx&itxʷ ‘do what to ?; do something to ’

gʷat ‘who?; someone’

ʔəx&ix&txʷyid ‘do what with  of ’s?; do something with  of ’s’

k’ʷid ‘how much/many?; some amount’

ʔidigʷat ‘say what?; say something’

liɬčad ‘which way?; some way’

ʔidigʷaac ‘say what to ?; say something to ’

pədtab ‘when?; sometime’

čad ‘where?; somewhere’

stab ‘what?; something’

čal ‘how? why?; somehow’

stabaɬ ‘what kind of ?; some kind of ’

čayɬ ‘go for what reason?; some reason for going’

tul’čad ‘from where?; from somewhere’

čədaɬ ‘which ?; some ’

Table 67: Interrogative/indefinite words

Five of the forms in this table (gʷat ‘who?’, stab ‘what?’, stabaɬ ‘what kind of?’, čədaɬ

‘which?’, pədtab ‘when?’) belong to the class of nouns, six are adverbs (čad ‘where?’, čal

‘how? why?’, dxʷčad ‘to where?’, k’ʷid ‘how much, how many?’, liɬčad ‘which way?’, tul’čad

‘from where?’), and the remaining six (ʔəx&id ‘what happened?’, ʔəx&itxʷ ‘do what to someone?’,

ʔəx&ix&txʷyid ‘do what with something of someone’s?’, ʔidigʷat ‘say what?’, ʔidigʷaac ‘say what

to someone?’) are verbs.

The forms in Table 67 are given with both their indefinite and interrogative glosses. As a matter of convenience,

the words will be glossed only as interrogatives for the remainder of the discussion.

249

Several of the forms in Table 67 are derivationally related to one of the other forms in the

table: pədtab ‘when?’ is formed by adding pəd- ‘seasonal’ (Section 2.2.6) to √tab ‘do’ (the latter

also being the base of stab ‘what?’) while stabaɬ ‘what kind?’ is derived from stab with the

incorporative suffix -aɬ (2.1.4). The question words dxʷčad ‘to where?’, liɬčad ‘which way?’,

and tul’čad ‘from where?’ are all derived directly by the combination of čad ‘where?’ with one

of the directional particles (2.7.2). Two of the words in Table 67, ʔəx&id ‘what happened?’ and

ʔidigʷat ‘say what?’, take one or more of the valency-increasing affixes: ʔidigʷat ‘say what?’ is

found with the allative applicative -c (2.1.2.6) in the form ʔidigʷaac ‘say what to someone?’;

ʔəx&id ‘what happened?’ takes the external causative suffix -txʷ (Section 2.1.2.2) to form ʔəx&itxʷ

‘do what to something?’. This form in turn takes the dative applicative -yi- (2.1.3.1), forming

ʔəx&ix&txʷyid ‘do what with something of someone’s?’. The specific uses of each of these question

words will be examined in turn, beginning with Section 2.6.1 below, following a discussion of

the properties common to all of the interrogative words.

In their most frequent use, interrogative words serve as the syntactic predicates of questions

that request information from the addressee about some person, thing, or event referred to by

their subject phrase:

(279) a. gʷat tiʔiɬ stubš

gʷat tiʔiɬ stubš

who DIST man

‘who is that man?’

(Bates, Hess & Hilbert 1994: 97)

b. gʷat kʷi ʔuʔəɬtxʷ čələp

gʷat kʷi ʔu–ʔəɬ–txʷ čələp

who REM PFV–eaten–ECS 2PL.SUB

‘who did you guys help?’

(Hess 1995: 100)

250

c. dxʷčad tiʔəʔ ʔuqadadid

dxʷ–čad tiʔəʔ ʔu–qada–di–d

CNTRPT–where PROX PFV–stolen–SS–ICS

‘where had those who had stolen from him gone?’

(Hess 2006: 57, line 379)

d. bəčayɬəxʷ čələp kʷi suʔux%ʷləp

bə=čayɬ=əxʷ čələp kʷi s=ʔu–ʔux%ʷ=ləp

ADD=go.for.what=now 2PL.SUB REM NM=PFV–go=2PL.PO

‘why did you guys (bother to) go?

(Hess 1998: 93, line 80)

The subject-phrases of interrogatives can be NPs (279a), headless relative clauses (279b) and (c),

or nominalizations (279d), depending on the syntactic relation of the questioned entity to the

head of the subject phrase. The syntax of information questions will be taken up in detail in

Section 8.4.2 below.

When used as clausal predicates, interrogative words that are not already verbs take on verbal

morphological and syntactic characteristics, including the potential to take aspectual prefixes:

(280) a. ʔəstab kʷi gʷədsq’p’ucid

ʔəs–stab kʷi gʷə=d=s=q’p’u–t–sid

STAT–what REM SBJ=1SG.PO=NM=pay–ICS–2SG.OBJ

‘what should I pay you?’

(Hess 2006: 30, line 190)

b. ləcuʔidigʷat čəxʷ

ləcu–ʔidigʷat čəxʷ

CONT–say.what 2SG.SUB

‘what are you saying?’

[MW Star Child, line 54]

c. ʔuʔəx%id əw’ə ʔə tiʔəʔ dəxʷcucuts

ʔu–ʔəx%id əw’ə ʔə tiʔəʔ dəxʷ=cut–cut=s

PFV–what.happen PTCL PR PROX ADNM=DSTR–say=3PO

‘how did she come to talk like that?’

(Hess 2006: 5, line 64)

Although any interrogative word used in this way may take aspectual prefixes, it is far more

common to find these affixes with those words that involve actions, motions, and states —

specifically, the interrogative adjectives čal ‘how?’, dxʷčad ‘to where?’, liɬčad ‘which way?’,

251

and tul’čad ‘from where?’, as well as the interrogative verbs čayɬ ‘go for what reason?’ ʔəx&id

‘what happened?’ and ʔidigʷat ‘say what?’ and their derivatives.

With the exception of čal

‘how?’, which seems to be in the process of fusing with the stative prefix ʔəs- (Section 2.6.5),

only the interrogative verbs appear to take aspectual inflections in their non-predicative uses.

In addition to serving as the syntactic predicates of information questions, interrogatives are

often used as indefinite anaphors with glosses like ‘something’ or ‘anywhere’, as shown in (281):

(281) a. gʷəl ʔuluudəxʷ tiʔiɬ stab

gʷəl ʔu–lu–d=əxʷ tiʔiɬ stab

SCONJ PFV–hear–ICS=now DIST what

‘then he heard something’

(Hess 1995: 148, line 24)

b. gʷəl xʷiʔəxʷ kʷi stabəxʷ gʷəšudub ʔə kʷi gʷəčad

gʷəl xʷiʔ=əxʷ kʷi stab=əxʷ gʷə=šuɬ–dxʷ–b ʔə kʷi gʷə=čad

SCONJ NEG=now REM what=now SBJ=see–DC–PASS PR REM SBJ=where

‘and nothing could be seen anywhere’

(Hess 2006: 27, line 118)

When used as indefinites, interrogative words behave syntactically as ordinary members of their

basic lexical class — that is, interrogative nouns behave as nouns, interrogative adverbs as

adverbs, and interrogative verbs as ordinary verbs. Because all three of these classes are potential

syntactic predicates, this makes it possible for interrogative words to appear in predicate position

in a clause with indefinite rather than interrogative meanings, giving rise to syntactically parallel

constructions with non-equivalent meanings:

(282) a. čad kʷi sʔibəš

čad kʷi s=ʔibəš=s

where REM NM=travel=3PO

‘he traveled everywhere’

(Hess 1998: 82, line 113)

Or, more accurately, it is more common to find these words inflected for aspects other than the imperfective,

which is a morphological zero and so would, by default, be the aspectual inflection of any interrogative word

lacking an overt aspectual prefix.

252

b. čad kʷi sʔibəš

čad kʷi s=ʔibəš=s

where REM NM=travel=3PO

‘where did he travel?’

In such indefinite uses, non-verbal interrogatives seem not to appear with aspectual inflection,

and all such expressions are distinguished from the analogous question through the use of

intonation (Hess, p.c.), although the exact nature of the intonational distinction (which is not

equivalent to the English-type sentence-final rise) is not fully understood at this point in time.

In their uses as indefinite anaphors, interrogatives frequently combine with the adverb bək’ʷ

‘all’ to create expressions equivalent to English everyone, everywhere, etc.:

(283) a. ləʔux%ʷcəxʷ kʷi bək’ʷ gʷat ʔal tiʔəʔ ʔəsɬaɬlil

lə=ʔux%ʷ–c=əxʷ kʷi bək’ʷ gʷat ʔal tiʔəʔ ʔəs–ɬaɬlil

PROG=go–ALTV=now REM all who at PROX STAT–live

‘[Changer] was going after everyone who lived there [in the world]’

[ML Mink and Tutyika I, line 171]

b. huy, q’ʷuʔtəbəxʷ tiʔiɬ ʔaciɬtalbixʷ tuˑl’ʔal bək’ʷ čad

huy q’ʷuʔtəbəxʷ tiʔiɬ ʔaciɬtalbixʷ tul’–ʔal bək’ʷ čad

SCONJ gathered–ICS–PASS=now DIST person CNTRFG–at all where

‘then the people were gathered from everywhere’

(Hess 1995: 142, line 47)

Similarly, in negative expressions interrogatives used as indefinites act like negative pronouns

such as nothing or no one:

(284) a. gʷəl xʷiʔ kʷi stab suʔəɬəds

gʷəl xʷiʔ kʷi stab s=ʔu–ʔəɬəd=s

SCONJ NEG REM what NM=PFV–feed.on=3PO

‘and they had nothing to eat’

(Hess 2006: 45, line 98)

b. gʷəl xʷiʔ gʷat ʔəsʔaladᶻiʔlyid

gʷəl xʷiʔ gʷat ʔəs–ʔaladᶻ•iʔl–yi–d

SCONJ NEG who STAT–care.for•child–DAT–ICS

‘but there was no one to babysit for her’

[MW Star Child, line 3]

253

Although interrogative words in such contexts are translated into English by negative lexical

items, their negative reading comes from context rather than from their actual semantics. The

syntax of negation is discussed in detail in Section 8.5 below.

Another use of interrogative words is that of introducing certain types of complement clause:

(285) a. laʔədəxʷ čad tiʔiɬ sʔa ʔə tiʔiɬ ləcuƛ’əladiʔ

laʔ–əd=əxʷ čad tiʔiɬ s=ʔa ʔə tiʔiɬ ləcu–ƛ’əladiʔ

locate–ICS=now where PROX NM=be.there PR DIST CONT–make.noise

‘he located where the sound was coming from’

[MW Star Child line 44]

b. gʷəl xʷiʔ kʷi gʷədəshaydxʷ stab tsiʔəʔ čəgʷas ʔə tiʔəʔ sgʷəlub

gʷəl xʷiʔ kʷi gʷə=d=s=ʔəs–hay–dxʷ stab tsiʔəʔ čəgʷas

SCONJ NEG REM SBJ=1SG.PO=NM=STAT–know–DC what PROX:FEM wife

ʔə tiʔəʔ sgʷəlub

PR PROX pheasant

‘but I do not know what the wife of Pheasant was’

(Hess 1998: 78, line 6)

In such contexts, the interrogative word introduces an embedded clause and serves as the

complement of a matrix verb, such as laʔəd ‘localize where’ in (285a) and haydxʷ ‘know that’ in

(285b), that can subcategorize for a sentential complement. For modern speakers, these clauses

take the form of an ordinary question. In more conservative style, however, interrogatives used

in these constructions act more like complementizers and take the subjunctive subject clitics:

(286) a. t’abad gʷəčadəs kʷi ɬusɬaɬlil čəɬ

t’abad gʷə=čad=əs kʷi ɬu=s=ɬaɬlil čəɬ

guess SBJ=where=3SBJ REM IRR=NM=live 1PL.PO

‘guess where we are going to live!’

(Hess 1995: ex. 1b)

b. xʷiʔ kʷi gʷədsəsaydxʷ gʷəstabəs kʷi ləgʷəčəd

xʷiʔ kʷi gʷə=d=s=ʔəs–hay–dxʷ gʷə=stab=əs kʷi lə=gʷəč–əd

NEG REM SBJ=1SG.PO=NM=STAT–know–DC SBJ=what=3SBJ REM PROG=search–ICS

‘I don’t know what he is looking for’

(Hess 1995: ex. 2b)

Sentences like these with embedded interrogative complementizers are discussed in more detail

in Section 9.4 below.

254

2.6.1 gʷat ‘who?’

Of the interrogative words, the most straightforward (from an English perspective, at any

rate) is the noun gʷat ‘who?; someone’, which is used in questions to ask for the identity of

humans and personified animals:

(287) a. gʷat čəxʷ

gʷat čəxʷ

who 2SG.SUB

‘who are you?’

b. gʷat tsiʔiɬ sɬadəyʔ

gʷat tsiʔiɬ sɬadəyʔ

who DIST:FEM woman

‘who is that woman?’

(Bates, Hess & Hilbert 1994: 97)

c. gʷatəxʷ kʷi ɬudsʔiɬugʷadadgʷad

gʷat=əxʷ kʷi ɬu=d=s=ʔiɬ–ʔu–gʷad–ad–gʷad

who=now REM IRR=1SG.PO=NM=PRTV–DSTR–DIM.EFF–converse

‘who will I converse with now?’

[DS Star Child, line 73]

As shown in (287), the subject phrase can be a subject-clitic, an NP, or a complex nominal

expression such as a headless relative clause.

When used in indefinite anaphoric expressions, gʷat is given the reading of ‘someone’:

(288) a. ɬuwiliq’ʷicid čəd dəbəɬ čəxʷ gʷat

ɬu=wiliq’ʷi–t–sid čəd dəbəɬ čəxʷ gʷat

IRR=query–ICS–2SG.OBJ 1SG.SUB belong.to.bloodline 2SG.SUB who

‘I will ask you [if] you are a descendant of someone’

[HM Star Child, line 114]

b. gʷəl kʷi gʷat sɬadəy’ ɬučəbaʔəd

gʷəl kʷi gʷat sɬadəy’ ɬu=čəbaʔ–əd

SCONJ REM who woman IRR=laden–ICS

‘some woman, she will bring it’

[DS Star Child, line 256]

(288a) shows gʷat acting as the predicate complement of the verb dəbəɬ ‘belong to a bloodline’,

which is itself the complement of wiliq’ʷid ‘query someone’. In (288b), gʷat is in a topicalized

subject phrase (Section *.*) and is modified by sɬadəy’ ‘woman’ to give the reading ‘some

255

woman’ — or, more literally, ‘someone [who is a] woman’. In negative expressions, ‘someone’

becomes ‘no one’:

(289) a. xʷiʔəxʷ kʷi gʷat gʷəʔux%

xʷiʔ=əxʷ kʷi gʷat gʷə=ʔux%

NEG=now REM who SBJ=go

‘there was not anyone to go’

[DS Star Child, line 301]

b. xʷiʔ kʷi gʷat gʷəhəliʔ

xʷiʔ kʷi gʷat gʷə=həliʔ

NEG REM who SBJ=alive

‘no one could survive’ (lit. ‘those who might live [are] not’)

(Hess 2006: 79, line 889)

gʷat is often found modified by the adverb bək’ʷ ‘all’ in expressions glossed as ‘everyone’:

(290) a. ləʔux%ʷcəxʷ kʷi bək’ʷ gʷat ʔal tiʔəʔ ʔəsɬaɬlil

lə=ʔux%ʷ–c=əxʷ kʷi bək’ʷ gʷat ʔal tiʔəʔ ʔəs–ɬaɬlil

PROG=go–ALTV=now REM all who at PROX STAT–live

‘[Changer] was going after everyone who lived there [in the world]’

[ML Mink and Tutyika I, line 171]

b. ƛ’asluutəb ʔə tiʔəʔ suʔululuɬ ʔə tiʔəʔ bək’ʷ gʷat ʔal kʷi səɬax%il

ƛ’u=ʔas–lu–t–əb ʔə tiʔəʔ s=ʔu–ʔul–uluɬ ʔə tiʔəʔ

HAB=STAt–heard–ICS–PASS PR PROX NM=PFV–DIM.EFF–go.by.water PR PROX

bək’ʷ gʷat ʔal kʷi s=lə=ɬax%–il=s

all who at REM NM=PROG=dark–INCH=3PO

‘they are heard by everyone paddling about in the evening’

(Hess 2006: 13, line 59a)

c. huy gʷəl, xʷiʔəxʷ gʷəsuʔitut ʔə tiʔəʔ bək’ʷ gʷat ʔal ti səsɬaɬlils

huy gʷəl xʷiʔ=əxʷ gʷə=s=ʔu–ʔitut ʔə tiʔəʔ bək’ʷ gʷat ʔal

SCONJ SCONJ NEG=now SBJ=NM=PFV–sleep PR PROX all who at

ti s=ʔəs–ɬaɬlil=s

NSPEC NM=STAT–live=3PO

‘and then no one could fall asleep while they lived there’

(Hess 2006: 3, line 14)

As shown in (290c), bək’ʷ gʷat is also glossed as ‘no one’ in negative expressions.

256

2.6.2 stab ‘what?’, stabaɬ ‘what kind of?’, and čədaɬ ‘which?’

The interrogative word stab ‘something; what?’ is derived from a combination of the verbal

radical √tab ‘do’ plus the nominalizing prefix s- (Section 2.2.1). It is used in questions to ask the

addressee to identify or name an object or non-human:

(291) a. stab tiʔiɬ

stab tiʔiɬ

what DIST

‘what is that?’

(Bates, Hess & Hilbert 1994: 216)

b. stab əw’ə dᶻəl tiʔiɬ ʔəswəliʔ šəq

stab əw’ə dᶻəl tiʔiɬ ʔəs–wəliʔ šəq

what PTCL PTCL DIST STAT–visible be.high

‘what is that up in the air?’

[DS Star Child, line 395]

c. ʔəstab kʷi gʷədsq’p’ucid

ʔəs–stab kʷi gʷə=d=s=q’p’u–t–sid

STAT–what REM SBJ=1SG.PO=NM=pay–ICS–2SG.OBJ

‘what should I pay you?’

(Hess 2006: 30, line 190)

When used as an indefinite anaphor, it can mean ‘something’, as in the examples in (292):

(292) a. gʷəl ʔuluudəxʷ tiʔiɬ stab

gʷəl ʔu–luu–d=əxʷ tiʔiɬ stab

SCONJ PFV–hear–ICS=NOW DIST what

‘he heard something’

(Hess 1995: 148, line 24)

b. xʷul’ čəxʷ ʔuhuyud ʔəsʔistaʔ stab kʷi adsdᶻaƛ’ad

xʷul’ čəxʷ ʔu–huyu–d ʔəs–ʔistaʔ stab kʷi

only 2SG.SUB PFV–be.done–ICS STAT–be.like what REM

ad=s=dᶻaƛ’a–d

2SG.PO=NM=confused–ICS

‘you just make it like something that confuses him’

(Hess 2006: 23, line 32)

In negative expressions, stab is glossed as ‘nothing’:

257

(293) a. xʷiʔ kʷi stab gʷətusuʔəɬəds

xʷiʔ kʷi stab gʷə=tu=s=ʔu–ʔəɬəd=s

NEG REM what SBJ=PAST=NM=PFV–feed.on=3PO

‘there was nothing they could eat’

(Hess 2006: 44, line 51)

b. xʷiʔ kʷi gʷəstab gʷəsəsč’x%idᶻ ʔal tiʔiɬ dəxʷʔas

xʷiʔ kʷi gʷə=stab gʷə=s=ʔəs–č’x%idᶻ ʔal tiʔiɬ dəxʷ=ʔa=s

NEG REM SBJ=what SBJ=NM=STAT–crowded at DIST ADNM=be.there=3PO

‘there was nothing that was crowded in there’

(Hess 2006: 67, line 619)

When combined with the adverb bək’ʷ ‘all’, stab forms an expression meaning ‘everything’:

(294) a. ɬusʔay’gʷasəxʷ ti bək’ʷ stab

ɬu=s=ʔay’•gʷas=əxʷ ti bək’ʷ stab

IRR=NM=changed•pair=now SPEC all what

‘everything will be changed now’

[ML Mink and Tutyika I, line 213]

b. tux%ʷ ƛ’astux%ʷtux%ʷ dəxʷɬid ʔə tiʔiɬ bək’ʷ stab, č’ic’qs, tiʔəʔ stab, tiʔəʔ x%ʷəx%ʷayuʔ, tiʔəʔ

bək’ʷ huyud titčulbixʷ

tux%ʷ ƛ’u=ʔas–tux%ʷ–tux%ʷ dəxʷ=ɬid ʔə tiʔiɬ bək’ʷ stab č’ic’qs

just IRR=STAT–DSTR–stretch ADNM=tied PR DIST all what mosquito

tiʔəʔ stab tiʔəʔ x%ʷəx%ʷayuʔ tiʔəʔ bək’ʷ huyu–d titčulbixʷ

PROX what PROX fly PROX all be.done–ICS small.animal

‘it[s web] would be stretched out so that everything is tied up, mosquitoes, things, flies,

all things that are made as small animals’

(Hess 2006: 30, line 208)

The phrase bək’ʷ stab can in turn be modified by a following noun (actually, a relativized

copular clause with a nominal predicate) that makes the expression more specific:

(295) a. qʷatqʷatatəbəxʷ tiʔəʔ bək’ʷ stab biac

qʷat–qʷata–t–əb=əxʷ tiʔəʔ bək’ʷ stab biac

DSTR–laid.out–ICS–PASS=now PROX all what meat

‘all the meat [lit. ‘all that was meat’] was laid out’

(Hess 1998: 72, line 187)

b. ƛ’ub čəɬ xʷuʔələʔ ʔuq’ʷuʔəd kʷi bək’ʷ stab titčul’bixʷ

ƛ’ub čəɬ xʷuʔələʔ ʔu–q’ʷuʔ–əd kʷi bək’ʷ stab titčul’bixʷ

well 1PL.SUB maybe PFV–gathered–ICS REM all what small.animal

‘we should, I guess, gather all kinds of [lit. ‘all that are’] small animals’

(Hess 2006: 18, line 159)

258

The effect of the following modifier in these examples is to narrow the scope of bək’ʷ stab from

‘every thing’ to ‘every thing that is an X’. This construction bears an obvious functional

relationship to the use of lexical suffixes illustrated in (302) below, and to the combination of

stab with the incorporative suffix -aɬ shown in (303).

Occasionally, stab turns up in texts where one might have expected gʷat ‘who?’, as in the

following examples:

(296) a. gʷəl huy, gʷadadgʷadəxʷ əlgʷəʔ gʷəstabəs kʷi ɬuc’əlalikʷ

gʷəl huy gʷadadgʷad=əxʷ əlgʷəʔ gʷə=stab=əs kʷi ɬu=c’əl–alikʷ

SCONJ SCONJ discuss=now PL SBJ=what=3SBJ REM IRR=win–ACT

‘and then they discussed who would win’

(Hess 1995: 144, line 31)

b. diɬəxʷ tudəxʷəsq’ʷuʔbitidəxʷ ʔə tiʔəʔ bək’ʷ stab: tiʔəʔ qaw’qs, tiʔəʔ k’aʔk’aʔ …

diɬ=əxʷ tu=dəxʷ=ʔəs–q’ʷuʔ–bi–t–id=əxʷ ʔə tiʔəʔ bək’ʷ

FOC=now PAST=ADNM=STAT–gathered–MAP–ICS–PASS.SBRD PR PROX all

stab tiʔəʔ qaw’qs tiʔəʔ k’aʔk’aʔ

what PROX raven PROX crow

‘that is why everyone is gathered together: Raven, Crow, …’

[DS Star Child, line 169]

However, in these cases (see also 285b above), both anaphoric expressions rather than questions,

the antecedents of stab are personified animals in traditional stories, which might explain the

storyteller’s use of the word, although gʷat can also be used in these contexts. There do seem to

be a handful of cases where the antecedents of stab are unequivocally human — although these,

too have alternative explanations:

(297) a. stab čəxʷ stab čəxʷ ʔukʷiˑxʷiˑd …

stab čəxʷ stab čəxʷ ʔu–kʷixʷid

what 2SG.SUB what 2SG.SUB PFV–pound

‘what are you, what are you who pounds?’

(Hess 1995: 141, line 37)

259

b. diɬ tsiʔəʔ adkiaʔ ʔi tsiʔəʔ adstab tsiʔəʔ ʔucucut

diɬ tsiʔəʔ ad–kiaʔ ʔi tsiʔəʔ ad–stab tsiʔəʔ

FOC PROX:FEM 2SG.PO–grandmother and PROX:FEM 2SG.PO–what PROX:FEM

ʔu–cut–cut

PFV–DSTR–say

‘that is your grandmother and your what’s-it-called who are speaking’

(Hess 2006: 69, line 645)

c. stab tiʔiɬ adsəskʷədiʔɬ

stab tiʔiɬ ad=s=ʔəs–kʷəd•iʔɬ

what DIST 2SG.PO=NM=STAT–held•child

‘what is that child you are holding?’

[DS Star Child, line 120]

In (297a), the import of stab may be a request for a name rather than an identity. The felicitous

answer to the question — directed to an old man working with an adze — would be a name or a

profession. In (297b), stab is used as a placeholder for a forgotten lexical item, so the anaphor

may indeed be an inanimate entity (the word sk’ʷuy ‘mother’) rather than the actual speaker (the

addressee’s mother).

Similarly, the aim of the question in (297c) is to find out the gender of the

child rather than its identity (cf. the English what is it, a boy or a girl?). Thus, while there may

seem to be some blurring between the uses of gʷat and stab for animates, there are few clear-cut

cases where stab is used for humans.

In addition to being glossed as ‘something’, stab in its non-interrogative uses often has the

sense of ‘thing’ or ‘unspecified object’:

(298) a. huy, ʔabyidəxʷ ʔə tiʔiɬ x%ʷuˑl’ paƛ’aƛ’ stab

huy ʔabyidəxʷ ʔə tiʔiɬ x%ʷul’ paƛ’aƛ’ stab

SCONJ extend–DAT–ICS=now PR DIST only worthless what

‘then he gave him just worthless things’

(Hess 1995: 147, line 17)

b. ƛ’al’ ʔəsƛ’altub ti ʔal tiʔiɬ adstab

ƛ’al’ ʔəs–ƛ’al–txʷ–b ti ʔal tiʔiɬ ad–stab

also STAT–covered–ECS–PASS SPEC at DIST 2SG.PO–what

‘also, he is covered in [i.e., wearing] your things’

(Hess 2006: 34, line 296)

In fact, stab or even stabəxʷ can be used as a “placeholder” while a speaker gropes for words.

260

This reading is found both when stab is used as a syntactic argument, as in (298), and when it is

used as a clausal predicate:

(299) a. haʔɬ stab

haʔɬ stab

good what

‘[it is a] good thing’

(Hess 2006: 38, line 396)

b. stab xʷuʔələʔ tiʔiɬ t’əbiɬəds

stab xʷuʔələʔ tiʔiɬ t’əbiɬəd–s

what maybe DIST rope–3PO

‘that thing is, I guess, his rope’ (lit. ‘his rope [is] maybe [that] thing’)

(Hess 2006: 29, line 178)

These uses of stab to mean ‘thing’ seem to be quite different from the indefinite uses illustrated

in (292), given that stab in all the expressions in (298) and (299) seems to refer to a definite

entity. This discrepancy might, however, be an artefact of translation stemming from the

differences between the way English and Lushootseed conceptualize the notion of indefiniteness.

(See the discussion of indefiniteness in determiners in Section 2.4 above.)

When modified by a relative clause, stab performs a function parallel to the English relative

pronouns what or that which:

(300) a. diɬ tsiʔəʔ ƛ’ustab tsiʔəʔ dəxʷəʔatəbəds əlgʷəʔ

diɬ tsiʔəʔ ƛ’u=stab tiʔəʔ dəxʷ=lə=ʔatəbəd=s əlgʷəʔ

FOC PROX:FEM HAB=what PROX ADNM=PROG=die=3PO PL

‘what is causing them to die [is] this’

(Hess 2006: 63, line 522)

b. ʔabyitəb ʔə kʷədiʔ stab suc’uqʷutəbs tul’ʔal tiʔəʔ cədiɬ stab sʔuladxʷ

ʔab–yi–t–əb ʔə kʷədiʔ stab s=ʔu–c’uqʷu–t–əb=s tul’–ʔal

extend–DAT–ICS PR REM.DMA what NM=PFV–sucked–ICS–PASS=3PO CNTRFG–at

tiʔəʔ cədiɬ stab sʔuladxʷ

PROX he what salmon

‘he was given what was [to be] sucked on by him from that which [was] a salmon’

(Hess 2006: 40, line 459)

Occasionally, stab is also found in such constructions acting as the main predicate of a clause:

David Beck 10-2-7 2:19 PM

Comment: this needs to be revised given the

shift from def to spec for this semantic

category

261

(301) a. diɬ stab gʷuhuyud

diɬ stab gʷə=ʔu–huyu–d

FOC what SBJ=PFV–be.done–ICS

‘[that is] what he made’

(Hess 2006: 42, line 13)

b. diɬəxʷ stab adsc’qib

diɬ=əxʷ stab ad=s=c’qib

FOC=now what 2SG.SUB=NM=get.a.share

‘[that is] what your share is’

(Hess 1998: 74, line 215)

The only two attestations of structures like this in the corpus make use of the focus particle, diɬ

(Section 11.2).

The interrogative stab ‘what?’ shows some potential to combine with lexical suffixes. The

resulting forms are interrogative words in which the suffixes act as classifiers, indicating the type

of item being asked about. In all, there are three attested forms of this type; however, it seems

possible that these are productive formations and stab can potentially combine with a much

wider range of lexical suffixes than are found in the present corpus. The forms that are attested

— stabaʔkʷəbixʷ ‘what group of people?’, stabac ‘what kind of tree?’, and stabidup ‘what

kind?’ — are illustrated in (302):

(302) a. stabaʔkʷəbixʷ čəxʷ

stab•aʔkʷə•bixʷ čəxʷ

what•group•cluster 2SG.SUB

‘what sort of person are you?’

b. stabac tiʔiɬ sƛ’ax%ʷdup

stab•ac tiʔiɬ s–ƛ’ax%ʷ•dup

what•tree DIST NP–grow•COLL

‘what kind of tree is that plant?

(Bates, Hess & Hilbert 1994: 216)

Included in this group might also be stabigʷs ‘possessions’, a lexicalized combination of stab and -igʷs, a lexical

suffix meaning ‘things, possessions’.

“A rather rude way of asking what group or tribe one belongs to” (Bates, Hess & Hilbert 1994: 216).

262

c. ʔux%ʷ tiʔiɬ stabidup, bird

ʔux%ʷ tiʔiɬ stab•idup bird

go DIST what•COLL bird

‘another kind, a bird, goes’

(Hess 2006: 19, line 186)

In each of these cases, the lexical suffix narrows the scope of the question to a particular kind of

entity that is being asked about.

In addition to combining with lexical suffixes, stab takes the incorporative suffix -aɬ

(Section 2.1.4), forming an new interrogative word, stabaɬ ‘what kind of something?’:

(303) a. stabaɬ titčulbixʷ

stab–aɬ titčulbixʷ

what–INCRP small.animal

‘what kind of little animal is that?’

b. stabaɬ əw’ə qʷɬayʔ tiʔiɬ ʔəsƛ’ax%ʷ ʔal tiʔiɬ

stab–aɬ əw’ə qʷɬayʔ tiʔiɬ ʔəs–ƛ’ax%ʷ ʔal tiʔiɬ

what–INCRP PTCL stick DIST STAT–grow at DIST

‘what kind of wood is that growing over there?

(Bates, Hess & Hilbert 1994: 216)

As with all incorporative stems, stabaɬ takes as a predicate complement a bare noun that serves

the same function as the lexical suffixes do in the examples in (302) — namely, to narrow the

scope of the question to a particular type of object.

A similar pattern is seen with the interrogative čədaɬ ‘which?’, which can also take a

nominal predicate complement, as in (304):

(304) čədaɬ sq’əlaɬəd kʷi ʔiɬx%aƛ’tx% čəxʷ

čədaɬ sq’əlaɬəd kʷi ʔiɬ–x%aƛ’–tx% čəxʷ

which berry REM PRTV–desire–ECS 2SG.SUB

‘which berry do you like best?’

(Bates, Hess & Hilbert 1994: 59)

As in the examples in (303), the predicate complement in this sentence, sq’əlaɬəd ‘berry’,

narrows the scope of the question by naming a type or set of objects from which the addressee

must select an individual entity. With čədaɬ, however, the presence of a nominal complement

appears to be optional:

263

(305) čədaɬ əw’ə kʷi hikʷ sq’əlaɬədac

čədaɬ əw’ə kʷi hikʷ sq’əlaɬəd•ac

which PTCL REM big berry•tree

‘which is the biggest berry bush?’

(Bates, Hess & Hilbert 1994: 59)

In (305), the speaker is simply requesting that the addressee point out a particular entity, the set

from which it is to be drawn being specified in the subject phrase, which in this case makes it

clear that the speaker is asking the addressee to single out a particular berry bush. Whether the

same possibility exists for stabaɬ or not remains an open question — there are currently no

attestations of stabaɬ used without a nominal predicate complement. The syntactic parallels

between stabaɬ and čədaɬ make it seem likely that, etymologically speaking, čədaɬ also contains

the incorporative suffix -aɬ — however, in the case of čədaɬ, there is no clear base for this

formation. Lacking further comparative evidence, this interrogative will have to be treated as

unanalyzable.

2.6.3 pədtab ‘when?’

Another word that may be derivationally related to stab is pədtab ‘when?’, formed with the

prefix pəd- ‘seasonal’ (Section 2.2.6). This interrogative is rather infrequent in texts, but follows

the same pattern as other interrogative words, acting as the predicate of a clause requesting

specific information from the addressee — in this case, a particular point in time described by

the clausal subject:

(306) a. pədtab kʷi ɬudšudubicid

pədtab kʷi ɬu=d=s=šuɬ–dxʷ–bicid

when REM IRR=1SG.PO=NM=see–DC–2SG.OBJ

‘when will I see you again?’

The Lushootseed Dictionary suggests that čədaɬ is derived from čad ‘where?’ and includes it as a sub-entry of the

latter. This seems possible, although the semantic shift from ‘where?’ to ‘which one (of something)?’ is

idiosyncratic enough that this etymology, if correct, is a deep historical artefact rather than a regular creation of

synchronic grammatical processes.

264

b. pədtab kʷi adəxʷəxʷcutəb gʷəsɬčils

100

pədtab kʷi ad=dəxʷ=ʔəs–dxʷ–cut–əb gʷə=s=ɬčil=s

when REM 2SG.PO=ADNM=STAT–CTD–say–DSD SBJ=NM=arrive=3PO

‘when do you think he will arrive?’

(Bates, Hess & Hilbert 1994: 216)

Etymologically, pədtab (usually reduced to [pətab]) may be derived from √tab ‘do’ rather than

directly from stab, although pəd- ‘seasonal’ is otherwise only attested as a nominal prefix.

2.6.4 čad ‘where?’, dxʷčad ‘to where?’, tul’čad ‘from where?’, and liɬčad ‘which way?’

The interrogative word čad ‘somewhere; where?’ is used to request information about the

location of an entity or event:

(307) a. tučadəxʷ čəxʷ

tu=čad=əxʷ čəxʷ

PAST=where=now 2SG.SUB

‘where have you been?’

(Bates, Hess & Hilbert 1994: 59)

b. čadəxʷ kʷi sbəq’ʷaʔ

čadəxʷ kʷi sbəq’ʷaʔ

where=now REM heron

‘where is Heron now?’

(Hess 2006: 19, line 177)

c. huy gʷəl čad kʷi sʔəy’dubs ʔə tiʔəʔ cədiɬ

huy gʷəl čad kʷi s=ʔəy’dxʷ–b=s ʔə tiʔəʔ cədiɬ

SCONJ SCONJ where REM NM=find–PASS=3PO PR PROX he

‘so then, where were they found by this one?’

(Hess 1998: 74, line 224)

(307) shows questions formed on the bare radical, čad, which is static and purely locative,

asking for the specific location of the referent of its subject. Like the locative-temporal

preposition ʔal (Section 2.3.1), čad is also frequently compounded with one of the directional

particles (2.7.2) to form more complex, dynamic stems that combine its basic locative meaning

with the notion of motion, direction, or point of origin:

100

The last word in this example is transcribed as gʷəsɬ cils in the original source.

265

(308) a. lədxʷčad čəxʷ

lə=dxʷ–čad čəxʷ

PROG=CNTRPT–where 2SG.SUB

‘where are you going to?

b. tul’čad čəxʷ

tul’–čad čəxʷ

CNTRFG–where 2SG.SUB

‘where are you coming from?’/‘where are you from?’

c. ƛ’uliɬčad čəxʷ

ƛ’u=liɬ–čad čəxʷ

HAB=PRLV–where 2SG.SUB

‘which way did you travel?’

(Bates, Hess & Hilbert 1994: 59)

In (308a), the combination of čad with the centripetal particle dxʷ questions the GOAL or the

point towards which the referent of the subject phrase is moving, the notion of motion being

implicit in the meaning of the directional particle. The sentence in (308b) illustrates much the

same point for the combination of čad and the centrifugal tul’, although in this case the motion

implicit in the directional particle could be interpreted as figurative, as the question is potentially

a question about where one is from in the sense of birthplace or former residence.

101

The final

example in (308c) shows the prolative particle used with čad; the result in this case is a

somewhat less transparent (though by no means idiosyncratic) interrogative word requesting

information about the subject’s path.

101

It is not clear if the same figurative interpretation is open to questions formed with the centripetal particle,

although there is one use of dxʷčad as an indefinite anaphor that seems to fit the bill:

(i) gʷič’gʷič’əxʷ dxʷʔal kʷi ɬudəxʷʔux%ʷs, dxʷčadəxʷ kʷi ɬuspaq’acuts tiʔəʔ ʔaciɬtalbixʷ

gʷič’gʷič’=əxʷ dxʷʔal kʷi ɬu=dəxʷ=ʔux%ʷ=s dxʷ–čad=əxʷ kʷi

search=now PR REM IRR=ADNM=go=3PO CNTRPT–where=now REM

ɬu=s=paq’a–t–sut=s tiʔəʔ ʔaciɬtalbixʷ

IRR=NM=distribute–ICS–REFL=3PO PROX person

‘they looked for where they could go, where the people could resettle’

(Hess 1998: 102, line 286)

The lack of questions where centripetal motion is figurative rather than literal may simply be the paucity of plausible

contexts for that type of expression.

266

As shown in the previous examples, when čad and the other locative interrogatives have

person-markers as subjects, they request information about the location of a particular entity. The

same is true of locative interrogatives that take NPs, headless relative clauses, or oblique-centred

nominalizations (Section 7.3) as subjects:

(309) a. čad kʷi x%ʷubt

čad kʷi x%ʷubt

where REM paddle

’where is the paddle?’

(Bates, Hess & Hilbert 1994: 59)

b. dxʷčad tiʔəʔ ʔuqadadid

dxʷ–čad tiʔəʔ ʔu–qada–di–d

CNTRPT–where PROX PFV–stolen–SS–ICS

‘where had those who had stolen from him gone?’

(Hess 2006: 57, line 379)

When the subject of the question is a sentential nominalization (Section 7.4.2.1), however, the

expression requests the location of a particular event or (in the case of a motion event) its path,

point of origin, or endpoint:

(310) a. čad kʷi ɬusgʷəƛ’əlad ʔə tiʔəʔ čaləs

čad kʷi ɬu=s=gʷəƛ’əlad ʔə tiʔəʔ čaləs–s

where REM IRR=NM=stop PR PROX hand–3PO

‘where will his arm stop?

(Hess 2006: 58, line 384)

b. tul’čad kʷi adskʷədxʷs tiʔiɬ adsʔəɬəd

tul’–čad kʷi ad=s=kʷəd–dxʷ tiʔiɬ ad–sʔəɬəd

CNTRFG–where REM 2SG.PO=NM=taken–DC DIST 2SG.PO–food

‘from where did you manage to get your food?’

(Hess 1998: 83, line 162)

c. liɬčad kʷi ɬadsuʔux%ʷ

liɬ–čad kʷi ɬu=ad=s=ʔu–ʔux%ʷ

PRLV–where REM IRR=2SG.PO=NM=PFV–go

‘which way are you going to go?’

(Bates, Hess & Hilbert 1994: 59)

267

d. tul’čad kʷi adsuʔibəš sgʷəlub

tul’–čad kʷi ad=s=ʔu–ʔibəš sgʷəlub

CNTRFG–where REM 2SG.PO=NM=PFV–go pheasant

‘where are you traveling from, Pheasant?’

(Hess 1998: 79, line 43)

Thus, in (310a) and (b), the speaker requests information about the location of an event, while in

(310c) the requested information is about path of motion, and (310d) questions point of origin.

Unlike the other interrogatives seen up until now, čad ‘where’ and its derivatives are

basically adverbs rather than nouns, and when used non-specifically they appear in the syntactic

position typical of locative adverbs, as shown in the examples in (311):

(311) a. ʔəbil’ ɬux%ʷil’ čad

ʔəbil’ ɬu=x%ʷil’ čad

if IRR=lost where

‘if he gets lost somewhere’

(Hess 2006: 23, line 31)

b. gʷəl ʔux%tubəxʷ ʔə kʷi stab dxʷčadəxʷ

gʷəl ʔux%–txʷ–b=əxʷ ʔə kʷi stab dxʷ–čad=əxʷ

SCONJ go–ECS–PASS=now PR REM what CNTRPT–where=now

‘and he is taken by something somewhere’

(Hess 2006: 27, line 121)

c. ʔəshaydxʷ dxʷčad

ʔəs–hay–dxʷ dxʷ–čad

STAT–know–DC CNTRPT–where

‘he knows which way to go towards’

(Hess 2006: 66, line 584)

In these expressions, the interrogative word appears following the main predicate of the clause,

in the position normally occupied by a locative adverb (2.5.2) or an adjunct (Section 8.2.7).

When used in this way, these words most frequently function as non-specific expressions.

However, there are examples where a word based on čad is sentence predicate and has a non-

specific rather than an interrogative reading:

268

(312) a. čad kʷi sʔibəš

čad kʷi s=ʔibəš=s

where REM NM=travel=3PO

‘he traveled everywhere’

(Hess 1998: 82, line 113)

b. dxʷčadəxʷ ʔal tiʔəʔ dəxʷgʷax%ʷs

dxʷ–čad=əxʷ ʔal tiʔəʔ dəxʷ=gʷax%ʷ=s

CNTRPT–where=now at PROX ADNM=walk=3PO

‘they walked all over’ (lit. ‘[they] went everywhere in their walking’)

(Hess 2006: 35, line 341)

This use of the locative interrogative words follows from their adverbial use shown in (311):

lexical and locative adverbs in Lushootseed are potentially clausal predicates (Sections 2.5.1 and

2.5.2), and so the sentences in (312) can be analyzed as clauses predicated on the (non-

interrogative) adverbial use of these words. The interrogative and non-interrogative uses of

locative interrogatives as clausal predicates is distinguished by intonation (Hess, p.c.).

Like the other interrogative words, the locative interrogatives combine with the adverb bək’ʷ

‘all’, in this case forming an expression meaning ‘everywhere’:

(313) a. ləʔibəš bək’ʷ dxʷčad

lə=ʔibəš bək’ʷ dxʷ–čad

PROG=travel all CNTRFG–where

‘he is traveling everywhere’

(Hess 1998: 65, line 14)

b. bək’ʷ čad ɬadəxʷutəl’awil ʔal taʔa qʷu, qəl’qəladiʔ

bək’ʷ čad ɬu=ad=dəxʷ=ʔu–təl’awil ʔal taʔa qʷuʔ qəl’–qəladiʔ

all where IRR=2SG.PO=ADNM–PFV–run at DIST:UNQ.DMA water DSTR–snag

‘you will be running everywhere by the water there, [in] the driftwood snags’

(Hess 1998: 74, line 239)

In both of these examples, the expressions bək’ʷ čad ‘everywhere’ and bək’ʷ dxʷčad ‘to

everywhere’ are used as indefinite adverbs.

The interrogative čad also combines with -il ‘inchoative’ to form the verb čadil ‘go off

somewhere’:

269

(314) x%ʷul’ ləčadil gʷəl lələk’ʷəd

x%ʷul’ lə=čad–il gʷəl lə=lək’ʷ–əd

only PROG=where–INCH SCONJ PROG=eaten–ICS

‘he was just going off somewhere and he was eating it’

(Hess 1998: 59, line 72)

Here, in spite of being the clausal predicate, čadil has a non-specific rather than an interrogative

reading, suggesting that this form, derived from an interrogative, may no longer itself be an

interrogative word. Unfortunately, there is only one attestation in the corpus, so it is unknown

whether or not the interrogative reading (i.e., ‘go off where?’) is open to this form as well.

2.6.5 čal ‘how?’

The interrogative word čal, glossed roughly as ‘how?’, is used to ask after the state or

condition of the entity referred to by its subject. When the subject is an NP, a headless relative

clause, or an oblique-centred nominalization, the question is interpreted as a request for

information about the current condition, status, or nature (as opposed to identity) of the subject:

(315) a. ƛ’asčaləxʷ čəxʷ

ƛ’u=ʔas–čal=əxʷ čəxʷ

HAB=STAT–how=now 2SG.SUB

‘how have you been?’

(Bates, Hess & Hilbert 1994: 60)

b. ʔəsčal əw’ə tiʔiɬ adsut’ilib

ʔəs–čal əw’ə tiʔiɬ ad=s=ʔu–t’ilib

STAT–how PTCL DIST 2SG.PO=NM=PFV–sing

‘how is your song?’

[DS Star Child, line 198]

In sentences like these, a more accurate English gloss of čal might be ‘like what?’, although

‘how?’ corresponds to the interrogative word used in most English translations of Lushootseed

sentences with čal. Note that, as in these examples, čal is almost invariably inflected for stative

aspect — indeed, it may be the case that the form ʔəsčal has become fossilized and has replaced

čal for many speakers.

270

Even more commonly than it takes NPs and complex nominal expressions such as those in

(315) as subjects, čal is found with a sentential nominalization (Section 7.4.2.1) as its subject in

a very regular construction used to request information about the manner in which something is

to be done or the means by which something is to be achieved:

(316) a. ɬasčaləxʷ kʷi ɬushuyud čəɬ

ɬu=ʔas–čal=əxʷ kʷi ɬu=s=huyu–d čəɬ

IRR=STAT–how=now REM IRR=NM=be.done–ICS 1PL.PO

‘how are we going to deal with it?’

(Hess 1998: 101, line 263)

b. gʷəl tux%ʷ ʔəsčal kʷi gʷadsuhuyuc

gʷəl tux%ʷ ʔəs–čal kʷi gʷə=ad=s=ʔu–huyu–t–s

SCONJ just STAT–how REM SBJ=2SG.SUB=NM=PFV–be.done–ICS–1SG.OBJ

‘but just how could you do it for me?’

(Hess 2006: 30, line 193)

The subject of these interrogative clauses is a nominalization based on the verbal radical √huy

‘be done’. Although this is a very specific type of construction, it accounts for a large proportion

of the attested uses of čal in the current corpus.

Additionally, čal is found in questions that ask about the manner in which a goal is to be

achieved. These questions take subject phrases formed with the adjunct nominalizer dəxʷ=

(Section 7.4.2.2), such as those in (317):

(317) a. ʔəsčal kʷi gʷədəxʷkʷədxʷs tsiʔiɬ bədaʔs

ʔəs–čal kʷi gʷə=dəxʷ=kʷəd–dxʷ=s tsiʔiɬ bədaʔ–s

STAT–how REM SBJ=ADNM=taken–DC=3PO DIST:FEM offspring–3PO

‘how would he manage to get his daughter?’

(Hess 1998: 95, line 133)

b. ʔəsčal kʷi gʷədəxʷlək’ʷdxʷyids tsiʔəʔ ʔalš ʔə tiʔəʔ sʔəɬəds

ʔəs–čal kʷi gʷə=dəxʷ=lək’ʷ–dxʷ–yi–d=s tsiʔəʔ

STAT–how REM SBJ=ADNM=eaten–DC–DAT–ICS=3PO PROX:FEM

ʔalš–s ʔə tiʔəʔ sʔəɬəd–s

cross.sex.sibling–3PO PR PROX food–3PO

‘how could he eat his sister’s food away from her?’

(Hess 1998: 56, line 6)

271

c. ʔəsčaləxʷ kʷi gʷədəxʷkʷədxʷ čəɬ

ʔəs–čal=əxʷ kʷi gʷə=dəxʷ=kʷəd–dxʷ čəɬ

STAT–how=now REM SBJ=ADNM=taken–DC 1PL.PO

‘how can we manage to obtain daylight?’

(Hilbert & Hess 1977: 13)

The use of dəxʷ=nominals — which are frequent in the expression of instrumental adverbial

adjuncts (Section *.*) — in sentences like these may be related to the implicit causal relation

between the desired outcome stated in the subject-phrase and the course of action being

questioned by the predicate. This is in contrast with the constructions in (316), which ask the

addressee to specify a course of action in response to a set of current conditions, the outcome of

which is unspecified.

When used as an indefinite anaphor, čal gives a reading of ‘however, in some way’:

(318) x%əɬ ti ʔubaʔscut ʔəsčal kʷədiʔ səshuys

x%əɬ ti ʔu–baʔs–t–sut ʔəs–čal kʷədiʔ s=ʔəs–huy=s

seemingly PFV–stationary–ICS–REFL STAT–how REM.DMA NM=STAT–be.done=3PO

‘he seemed to be stationary, however he did that’

(Hess 2006: 55, line 316)

The indefinite use of čal is occasionally found in predicative uses of the word:

(319) a. ʔəsčaləxʷ kʷi tushuyutubs

ʔəs–čal=əxʷ kʷi tu=s=huyu–txʷ–b=s

STAT–how=now REM PAST=NM=be.done–ECS–PASS=now

‘or whatever it was he had done to her’ (lit. ‘what he did to her [was] in some way’)

(Hess 2006: 21, line 224)

b. ƛ’asčal diʔaʔəxʷ kʷaʔ

ƛ’u=ʔas–čal diʔaʔ=əxʷ kʷaʔ

HAB=STAT–how here=now PTCL

‘they have been around here, though’

(Hess 2006: 44, line 63)

Neither of these uses of čal is particularly common in the present corpus, although they are

predicted by analogy with the other interrogative words, which would lead us to expect that čal

is basically an indefinite adverb that, like čad, can appear in predicate position in interrogative

and non-interrogative expressions, the two types of sentence being distinguished by intonation.

čal is found with the nominalizing prefix s-, forming the lexical item sčal ‘way, manner’:

272

(320) xʷiʔəxʷ gʷəstabəxʷ gʷəsʔidigʷatəxʷ, gʷəstab, gʷəsčal

xʷiʔ=əxʷ gʷə=stab=əxʷ gʷə=s–ʔidigʷat=əxʷ gʷə=stab gʷə=s–čal

NEG=now SBJ=what=now SBJ=NP–say.what=now SBJ=what SBJ=NP–how

‘nothing was said, nothing in any way’

(Hess 1998: 101, line 282)

With the exception of the lexicalized expression sk’ʷid ‘time, hour’ given in (326) below, the

analogous forms for the other adverbial interrogatives are not attested in the present corpus.

2.6.6 k’ʷid ‘how many? how much?’

The interrogative word k’ʷid ‘some amount; how many, how much?’ requests information

about the number or quantity of items specified by the subject phrase:

(321) a. k’ʷid kʷi adsʔuladxʷ

k’ʷid kʷi ad–sʔuladxʷ

how.many REM 2SG.PO–salmon

‘how many salmon do you have?’

(Bates, Hess & Hilbert 1994: 131)

b. k’ʷid kʷədiʔ səsʔahačiʔs

k’ʷid kʷədiʔ s=ʔəs–ʔa•hačiʔ=s

how.many REM.DMA NM=STAT–be.there•hand=3PO

‘how many did they put their hands to [i.e., heal] there?’

(Hess 2006: 63, line 506)

c. ƛ’ucutəb, k’ʷid sq’aʔšəd

ƛ’u=cut–t–əb k’ʷid sq’aʔšəd

HAB=say–ICS–PASS how.many moccasin

‘he was asked, “how many moccasins?”’

[DS Star Child, line 350]

As with the other interrogatives, the subject phrase of questions based on k’ʷid can be an

ordinary NP (321a) or a complex nominal expression (321b). As shown in (321c), k’ʷid can also

take a nominal predicate complement, narrowing the scope of the inquiry by specifying what

type of entity the addressee is being asked to quantify. The formulation given in (321a) is the

most common way of asking ‘how much (of something) do you have?’.

Like stab ‘what?’, k’ʷid is regularly found associated with lexical suffixes. In these

constructions, the lexical suffixes act as classifiers, indicating the type of item being asked about:

273

(322) a. k’ʷidilc kʷi adtalə

k’ʷid•ilc kʷi ad–talə

how.many•round.object REM 2SG.PO–dollar

‘how much money do you have?’

b. k’ʷidalq čəxʷ

k’ʷid•alq čəxʷ

how.many•game 2SG.SUB

‘how much game do you have?’

(Bates, Hess & Hilbert 1994: 131)

While stab is attested with only a few lexical suffixes, k’ʷid combines with a much wider

(perhaps unrestricted) range: in the present corpus, there are attestations of k’ʷidalps ‘how many

(economically important) animals?’, k’ʷidaʔltxʷ ‘how many houses?’, k’ʷidalq ‘how much

game?’, k’ʷidgʷiɬ ‘how many canoes?’, and k’ʷidilc ‘how much money?’.

102

Unlike stab, k’ʷid does not combine on its own with the incorporative suffix -aɬ, but does

participate in compounds combining -aɬ and some other element, either a lexical suffix or an

independent noun:

(323) a. xʷiʔəxʷ kʷi stabəxʷ gʷəƛ’(u)a(s)šudxʷ əlgʷəʔ dxʷʔal kʷədiʔ tuk’ʷidəɬdat

xʷiʔ=əxʷ kʷi stab=əxʷ gʷə=ƛ’u=ʔas–šuɬ–dxʷ əlgʷəʔ dxʷ–ʔal

NEG=now REM what=now SBJ=HAB=STAT–see–DC PL CNTRPT–at

kʷədiʔ tu=k’ʷid–əɬ•dat

REM.DMA PAST=how.many–INCRP•days

‘they could not see things for many days’

(Hess 2006: 53, line 263)

b. k’ʷidəɬtaɬ

k’ʷid–əɬ–taɬ

how.many–INCRP–fathom

‘how many fathoms [is it]?’

(Bates, Hess & Hilbert 1994: 131)

c. k’ʷidəladxʷəxʷ tiʔiɬ stubš

k’ʷid–əl•adxʷ=əxʷ tiʔiɬ stubš

how.many–INCRP•year=now DIST man

‘how old is that man?’ (lit. ‘how many years is that man?’)

(Bates, Hess & Hilbert 1994: 131)

102

The Lushootseed Dictionary also contains the form k’ʷidwač ‘what time?’, based on the English watch.

274

d. huy gʷəl x%ʷul’əxʷ x%əɬ ti x%ʷul’ k’ʷidəɬslək’ʷ tiʔiɬ slək’ʷtəbs

huy gʷəl x%ʷul’=əxʷ x%əɬ ti x%ʷul’ k’ʷid–əɬ–s–lək’ʷ tiʔiɬ

SCONJ SCONJ only=now seemingly only how.many–INCRP–NP–eaten DIST

s=lək’ʷ–t–əb=s

NM=eaten–ICS–PASS=3PO

‘and then it was just as though in just a few gulps Raven ate it [all]’

(Hess 1998: 84, line 167)

In (323a), k’ʷid combines with -aɬ and the lexical suffix -dat ‘day’ to form an expression ‘how

many days?’, used in this example as an indefinite anaphor. In (323b), it combines with the noun

taɬ ‘fathom’ (more precisely, a measure from fingertip to fingertip with one’s arms outstretched),

and in (323c) it forms a compound with the lexical nominalization slək’ʷ, which presumably

means ‘swallow’ or ‘gulp’ (cf. lək’ʷəd ‘eat something, put something in mouth’), although it has

no other attestations in the corpus or the dictionary. It is not clear to what extent these

compounds are lexicalized constructions or if they are in fact productive uses of k’ʷid to form

novel compound interrogative forms. Note also that the use of the incorporative suffix -aɬ with a

lexical suffix like -dat in (323a) or -adxʷ in (323c) is not attested in other constructions, and is

suggestive of a diachronic path for lexical suffixes from independent nouns (the usual type of

predicate complement selected by -aɬ) to bound suffix.

Like many words associated with numeration and counting, k’ʷid has a special form used to

ask about the number of people:

(324) ʔəbsbibədbədaʔ ʔə kʷi tuk’ʷidid

103

ʔəs–bəs–bi–bəd–bədaʔ ʔə kʷi tu=k’ʷidid

STAT–PROP–ATTN–DSTR–child PR REM PAST=how.many:HMN

‘she had how many little children?’

(Hess 2006: 43, line 36)

This form of the interrogative, k’ʷidid, is created by Type III plural reduplication (Section 5.3.4).

It is only used when asking about third-persons:

103

Note that this construction also appears to be an example of an in situ question, a construction otherwise

unattested in the present corpus. The fact that it is indeed an interrogative rather than an indefinite expression is

marked by intonation (Hess, p.c.)

275

(325) k’ʷid čələp

k’ʷid čələp

how.many 2PL.SUB

‘how many of you?’

(Bates, Hess & Hilbert 1994: 131)

When enquiring about the number of second persons, the singular form of the interrogative,

k’ʷid, is used, as shown in (325).

Unlike most of the other interrogative words, k’ʷid has some idiomatic uses that are not

entirely expected from its English gloss. One of these has to do with asking about the time,

which can be done with the expression ʔaləxʷ sk’ʷid, as in the following examples:

(326) a. ʔaləxʷ sk’ʷids

ʔal=əxʷ s=k’ʷid

PR=now NM=how.many

‘what time is it?’

b. ʔəsaydxʷ čəxʷ ʔu ʔaləxʷ sk’ʷids kʷi sʔux%ʷs

ʔəs–hay–dxʷ čəxʷ ʔu ʔal=əxʷ s–k’ʷid–s kʷi s=ʔux%ʷ=s

STAT–know–DC 2SG.SUB INT PR=now NP–how.many–3PO REM NM=go=3PO

‘do you know what time he went?’

(Bates, Hess & Hilbert 1994: 130)

sk’ʷid is derived from the combination of k’ʷid with the nominalizing prefix s-. Compositionally

this form would be expected to mean something like ‘amount’ or ‘quantity’, but appears to have

been lexicalized to mean ‘time’. This form is only attested as part of the fixed expression ʔaləxʷ

sk’ʷid which, as shown in (326), has both interrogative and indefinite readings.

The interrogative k’ʷid also appears in the expression x&ʷul’ k’ʷid ‘not many, only a few’ (lit.

‘only how many?, only many’):

(327) gʷəl tiʔəʔ sp’ic’ikʷ gʷəl x%ʷul’ k’ʷid sləx%il gʷəl luƛ’əxʷ stubš

gʷəl tiʔəʔ sp’ic’ikʷ gʷəl x%ʷul’ k’ʷid sləx%il gʷəl luƛ’=əxʷ stubš

SCONJ PROX Diaper.Child SCONJ only how.much day SCONJ old=now man

‘as for Diaper Child, it was not many days [before] he grew into a man’

[DS Star Child, line 172]

This seems to be the opposite to the expectation created by the literal English gloss, according to

which the interpretation is more likely to be ‘many’ rather than ‘only a few’.

276

2.6.7 čayɬ ‘go for what reason?’ and ʔidigʷat ‘say what?’

The interrogative verbs, čayɬ ‘go for what reason?’ and ʔidigʷat ‘say what?’, request

information about a particular type of event whose nature is specified in the meaning of the

interrogative itself. The first of these, čayɬ ‘go for what?’, asks for the motive behind someone’s

going somewhere:

(328) a. čayɬ čəxʷ

čayɬ čəxʷ

go.for.what 2SG.SUB

‘why do you want to go?’

b. ləčayɬəxʷ kʷi sʔux%ʷs ʔal tiʔiɬ ʔuq’ʷuʔq’ʷuʔ

lə=čayɬ=əxʷ kʷi s=ʔux%ʷ=s ʔal tiʔiɬ ʔuq’ʷuʔq’ʷuʔ

PROG=go.for.what=now REM NM=go=3PO at DIST gathering

‘why should she want to go to that gathering?

(Bates, Hess & Hilbert 1994: 61)

The subject phrase of a question based on čayɬ can be an NP, a pronoun, or a subject-clitic (as in

328a) or it can be a sentential nominalization, as in (328b). In the latter case, the notion of

‘going’ is stated overtly in the subject phrase; in the former, it is expressed only as part of the

meaning of the interrogative word itself. In the current corpus, čayɬ is infrequent and appears

only in interrogative contexts.

The second of these two verbs, ʔidigʷat ‘say what?’, requests information about the contents

of a speech or communicative act:

(329) a. ʔuʔidigʷat čəxʷ

ʔu–ʔidigʷat čəxʷ

PFV–say.what 2SG.SUB

‘what did you say?’

(Bates, Hess & Hilbert 1994: 15)

b. ʔuˑ, ləcuʔidigʷat čələp, wiw’su

ʔu ləcu–ʔidigʷat čələp wiw’su

INTJ CONT–say.what 2SG.SUB children

‘oh, what are you saying, children?’

[AJ Basket Ogress, line 95]

277

c. ʔəsʔidigʷat ti syəcəb ʔal tiʔiɬ sx%al

ʔəs–ʔidigʷat ti s–yəcəb ʔal tiʔiɬ s–x%al

STAT–say.what SPEC NP–report at DIST NP–write

‘what is reported in this document?’

(Bates, Hess & Hilbert 1994: 16)

As with čayɬ, ʔidigʷat includes the event-type (usually, ‘say’) in its own meaning, and does not

require that it be expressed overtly by the subject-phrase, although this is an option, as in (329c).

When used in non-interrogative contexts, ʔidigʷat acts as an indefinite anaphoric expression

meaning ‘say something’:

(330) a. xʷiʔ gʷəƛ’usuʔidigʷat ʔə tsiʔəʔ waq’waq’

xʷiʔ gʷə=ƛ’u=s=ʔu–ʔidigʷat ʔə tsiʔəʔ waq’waq’

NEG SBJ=HAB=NM=PFV–say.what PR PROX.FEM frog

‘Frog would not say anything (intelligible)’

(Hess 2006: 3, line 17)

b. x%əɬ tihəxʷ xʷuʔələʔ gʷələliʔluud əlgʷəʔ kʷi gʷəsuʔidigʷat ʔə tiʔəʔ qʷiqʷqʷistay’bixʷ

x%əɬ ti=həxʷ xʷuʔələʔ gʷə=lə=liʔ–luu–d əlgʷəʔ kʷi

seemingly=now maybe SBJ=PROG=ATTN–hear–ICS PL REM

gʷə=s=ʔu–ʔidigʷat ʔə tiʔəʔ qʷi–qʷ–qʷistay’bixʷ

SBJ=NM=PFV–say.what PR PROX ATTN–ATTN–dwarf

‘it seems maybe they were understanding a bit of whatever these little dwarves said’

(Hess 2006: 64, line 533)

Most of the attested uses of this type are in negative expressions like that in (330a), which is the

usual way of expressing ‘not say anything’.

As a verb, ʔidigʷat takes the nominalizing prefix s- to form a word sʔidigʷat ‘something

said, statement’:

(331) xʷiʔəxʷ gʷəstabəxʷ gʷəsʔidigʷatəxʷ, gʷəstab, gʷəsčal

xʷiʔ=əxʷ gʷə=stab=əxʷ gʷə=s–ʔidigʷat=əxʷ gʷə=stab gʷə=s–čal

NEG=now SBJ=what=now SBJ=NP–say.what=now SBJ=what SBJ=NP–how

‘nothing was said, nothing in any way’

(Hess 1998: 101, line 282)

The analogous form for čayɬ, sčayɬ ‘some reason for going’, is predicted to exist, but is not

attested in the present corpus.

278

As noted above, ʔidigʷat combines with the allative applicative -c to form ʔidigʷaac ‘say

what to someone?’:

(332) xʷiʔ kʷi ɬadsʔidigʷaac tsiʔiɬ adsk’ʷuy ɬukʷədatəbəs tiʔiɬ adbədaʔ

xʷiʔ kʷi ɬu=ad=s=ʔidigʷat–c tsiʔiɬ ad–sk’ʷuy

NEG REM IRR=2SG.PO=NM=say.what–ALTV DIST:FEM 2SG.PO–mother

ɬu=kʷəda–t–əb=əs tiʔiɬ ad–bədaʔ

IRR=taken–ICS–PASS=3SBJ DIST 2SG.PO–offspring

‘do not say anything to your mother when she takes your son’

(Bates, Hess & Hilbert 1994: 16)

Unfortunately, this word is not attested in questions, but instead appears in the available data

only in indefinite anaphoric uses such as that shown here in (332). It seems likely, based on

analogy with other interrogative words that take valency-increasing affixes, that both

interrogative and anaphoric uses are open to this word, although resolution of this question will

depend on uncovering further examples.

2.6.8 ʔəx%id ‘what happened?’

In questions, the interrogative verb ʔəx&id ‘what happened?’ is used to inquire in a general

way about an action or event, the exact nature of the information being requested depending on

the form of the question itself. In questions with simple NP subjects, the information requested is

a specification of an event in which the entity identified by the subject phrase was a participant:

(333) a. ʔuʔəx%id əw’ə čəxʷ

ʔu–ʔəx%id əw’ə čəxʷ

PFV–what.happen PTCL 2SG.SUB

‘what are you up to?’

(Bates, Hess & Hilbert 1994: 13)

b. gʷəʔəx%idəxʷ čələp gʷəɬčisəbaləp ʔə kʷsi ʔax%ʷadus

gʷə=ʔəx%id=əxʷ čələp gʷə=ɬčil–s–əb=aləp ʔə kʷsi

SBJ=what.happen=now 2PL.SUB SBJ=arrive–ALTV–PASS=2PL.SBJ PR REM:FEM

ʔax%ʷadus

Basket.Ogress

‘what will you guys do if you are come upon by the Basket Ogress?’

[DM Basket Ogress, line 6]

279

In these examples, ʔəx&id is used to ask about an event as identified by its AGENT. In these

contexts, the best approximation of its meaning is ‘do what?’, where the scope of the

interrogative encompasses an entire event which is identified for the addressee by the identity of

one of its participants (in 333, the AGENT or ACTOR). It is also possible to ask about an event

identified by its PATIENTs or UNDERGOERs, as in (334):

(334) a. ʔuʔəx%id kʷi kikəwič

ʔu–ʔəx%id kʷi ki–kəwič

PFV–what.happen REM ATTN–hunchback

‘what happened to Little Hunchback?’

[DM Basket Ogress, line 79]

b. ʔəx%id tiʔəʔ bədaʔs əlgʷəʔ kʷədiʔ tudəxʷʔatəbəds

ʔəx%id tiʔəʔ bədaʔ–s əlgʷəʔ kʷədiʔ tu=dəxʷ=ʔatəbəd=s

what.happen PROX child–3PO PL REM.DMA PAST=ADNM=die=3PO

‘what happened to their son that he died?’

[ML Mink and Tutyika I, line 44]

c. ʔuʔəx%id čələp, ʔuʔəx%id

ʔu–ʔəx%id čələp ʔu–ʔəx%id

PFV–what.happen 2SG.SUB PFV–2SG.SUB

‘what happened to you guys? what happened?’

[DM Basket Ogress, line 77]

In sentences like these, the most straightforward English gloss would be ‘what happened to?’;

however, syntactically, the questions in (333) and (334) are identical and the apparent

differences between them is an artefact of translation. A more literal rendition of questions of

this type would be something like ‘what was it that happened involving X?’, X’s role as AGENT

or PATIENT being irrelevant to the syntax of the question. Indeed, ʔəx&id is often used to ask about

events in which the referent of the subject phrase may not have had a direct role at all, as in the

examples in (335), where the interrogative might best be glossed as ‘what is the matter with?’

and asks about the general circumstances surrounding or affecting someone:

280

(335) a. ʔuˑ, ʔuʔəx%id čələp

ʔu ʔu–ʔəx%id čələp

INTJ PFV–what.happen 2PL.SUB

oh, what’s the matter with you guys?’

[AJ Basket Ogress, line 82]

b. ʔəsʔəx%id əw’ə tsiʔəʔ adčəgʷas dəx%ʷul’s ʔubak’ʷacut tiʔəʔ qədxʷs

ʔəs–ʔəx%id əw’ə tsiʔəʔ ad–čəgʷas dəxʷ=x%ʷul’=s

STAT–what.happen PTCL PROX:FEM 2PO–wife ADNM=only=3PO

ʔu–bak’ʷa–t–sut tiʔəʔ qədxʷ–s

PFV–move.quickly–ICS–REFL PROX mouth–3PO

‘what is the matter with your wife that her mouth is just a-goin’?

(Hess 2006: 4, line 22)

c. ʔəsʔəx%idəxʷ kʷi ʔaciɬtalbixʷ gʷasʔitutəs

ʔəs–ʔəx%id=əxʷ kʷi ʔaciɬtalbixʷ gʷə=ʔas–ʔitut=əs

STAT–what.happen=now REM people SBJ=STAT–sleep=3SBJ

‘why are the people asleep?’

(lit. ‘what is the matter with the people that they are asleep?’)

(Hilbert & Hess 1977: 30)

As in (333), the subject of the interrogative word in these constructions is the affected

participant, but — unlike (333) — the appropriate answer to the questions in (335) may or may

not require that participant to be the subject of the sentence.

When the subject phrase of the question is a sentential nominalization referring to an event

(Section 7.3), ʔəx&id is interpreted as asking in a general way for further information about that

event with glosses like ‘why?’, ‘how far?’, or ‘how long?’:

(336) a. bəʔəx%idəxʷ kʷi adəxʷuxʷəcəd kʷi adsƛ’alabac

bə=ʔəx%id=əxʷ kʷi ad=dəxʷ=ʔu–xʷəc–əd kʷi

ADD=what.happen=now REM 2SG.PO=ADNM=PFV–remove–ICS REM

ad=s=ƛ’al•abac

2SG.PO=NM=cover•body

‘why are you taking your clothes off again?’

(Hess 2006: 35, line 323)

b. ʔəsʔəx%id kʷi sʔibəš dxʷʔaɬx%adulgʷədxʷ

ʔəs–ʔəx%id kʷi s=ʔibəš=s dxʷ–ʔaɬx%adulgʷədxʷ

STAT–what.happen REM NM=travel=3PO CNTRFG–downstream

how far did he travel downstream?’

[MW Star Child, line 37]

281

c. x%ʷul’əxʷ ʔəsʔəx%id tiʔiɬ sʔa ʔə cədiɬ tučəgʷas dxʷʔal t’aq’t

x%ʷul’=əxʷ ʔəs–ʔəx%id tiʔiɬ s=ʔa ʔə cədiɬ tu=čəgʷas–s

only=now STAT–what.happen DIST NM=exist PR he PAST=wife–3PO

dxʷ–ʔal t’aq’t

CNTRPT–at inland

‘just how long was his wife there up from shore?’

(Hess 2006: 16, line 126)

d. ʔəsʔəx%id kʷi gʷadscutəbš dxʷʔal kʷi gʷəcəxʷhuyucid ʔəsʔistaʔ

ʔəs–ʔəx%id kʷi gʷə=ad=s=cut–t–əbš dxʷʔal kʷi

STAT–what.happen REM SBJ=2SG.PO=NM=say–ICS–1SG.OBJ CNTRPT–at REM

gʷə=d=dəxʷ=huyu–t–sid ʔəs–ʔistaʔ

SBJ=1SG.PO=ADNM=be.done–ICS–2SG.OBJ STAT–be.like

‘what are you saying to me that I should treat you that way?’

(Hess 2006: 29, line 165)

If the event in the subject phrase is vague or expressed by a generic verb like huy ‘be done’,

ʔəx&id can also be interpreted (more along the lines of 333 and 334) as a request for information

about the nature of the event itself:

(337) ʔəsʔəx%id, ʔu dsuq’ʷsuq’ʷaʔ, kʷi gʷədshuy ʔal ti

ʔəs–ʔəx%id ʔu d–suq’ʷ–suq’ʷaʔ kʷi gʷə=d=s=huy

STAT–what.happen INT 1SG.PO–DSTR–younger.sibling REM SBJ=1SG.PO=NM=be.done

ʔal ti

at SPEC

‘what, my little brothers, can I do about this?’

(Hess 2006: 23, line 25)

Thus, the exact relationship between the subject phrase and the information being requested

seems to be inferred from context rather than being specified exactly by the interrogative as it is

in the English translations.

In at least one example in the corpus, ʔəx&id appears to take a predicate complement:

(338) ʔəsʔəx%id gʷəscutəbs

ʔəs–ʔəx%id gʷə=s=cut–t–əb=s

STAT– what.happen SBJ=NM=say–ICS–PASS=3PO

‘what would it be called?’

(Hess 2006: 7, line 113)

282

Here, the nominalized passive scutəbs ‘his/her saying to it’ is part of the predicate phrase headed

by ʔəx&id and serves to narrow the scope of the question by specifying that the request is for

information about an event of speaking (or, in this case, of naming something). The contrast

between this type of construction and those shown in (336) and (337) in which the event is

specified in the subject phrase probably resides in the fact that in the latter cases the event being

referred to is an actual, single (and topical) event known both to the speaker and the addressee,

whereas in (338) the event specified by the predicate complement is a generic or habitual action.

Comparative questions soliciting the difference between two things are also based on ʔəx&id:

(339) ʔəsʔəx%id kʷi sləliʔ ʔə tiʔiɬ haʔac dxʷʔal sx%əp’ab

ʔəs–ʔəx%id kʷi s=ləliʔ ʔə tiʔiɬ haʔac dxʷ–ʔal sx%əp’ab

STAT–what.happen REM NM=differ PR DIST horse.clam CNTRPT–at cockle

‘how does a horse clam differ from a cockle?’

(Hess & Hilbert 1976: II, 74)

Comparative constructions are discussed in more detail in Section 8.8.

When used as an indefinite anaphor, ʔəx&id means something along the lines of ‘what

happens’ or ‘that which takes place’, as in (340):

(340) a. gʷəl diɬ sʔa kʷi ɬusʔəx%id ʔə tiʔiɬ adbədaʔ

gʷəl diɬ s=ʔa kʷi ɬu=s=ʔəx%id ʔə tiʔiɬ ad–bədaʔ

SCONJ FOC NM=be.there REM IRR=NM=what.happen PR PROX 2SG.PO–offspring

‘and your son will be born there’

(lit. ‘and it is your son’s being there that will be what happens’)

[DS Star Child, line 108]

b. xʷiʔ gʷəƛ’usuyəcəbs ʔəsʔəx%id tiʔəʔ šəgʷɬ tiʔiɬə dəxʷaʔɬs

xʷiʔ gʷə=ƛ’u=s=ʔu–yəcəb=s ʔəs–ʔəx%id tiʔəʔ šəgʷɬ tiʔiɬə

NEG SBJ=HAB=NM=PFV–report=3PO STAT–what.happen PROX path DIST

dəxʷ=haʔɬ=s

ADNM=good=3PO

‘it is not said what happens on the road where it is good’

[AW Basket Ogress, line 8]

283

c. stab kʷi gʷəƛ’usʔəx%id čəɬ

stab kʷi gʷə=ƛ’u=s=ʔəx%id čəɬ

what REM SBJ=HAB=NM=what.happen 1SG.PO

‘what could we do?’ (lit. ‘what we could do [is] what?’)

(Hess 2006: 45, line 93)

Note that in (340c) the subject phrase is based on a clause, ʔəx&id čəɬ, which could be translated

variously as ‘what did we do?’ (like the analogous clauses in 333), ‘what happened to us?’ (as in

334), or even (as in 335) ‘what is the matter with us?’. As an indefinite anaphor, this phrase in

isolation would be glossed something along the lines of ‘that which happened involving us’, the

specific semantic role played by ‘us’ being left to context to clarify.

When modified by the adverb bək’ʷ ‘all’, ʔəx&id forms part of an expression meaning ‘in

every way’:

(341) a. bək’ʷ ʔəsʔəx%id haʔɬ šuɬ …

bək’ʷ ʔəsʔəx%id haʔɬ šuɬ

all STAT–what.happen good see

‘he is good-looking in every way’

(Bierwert 1996: 127, line 206)

b. bək’ʷ ʔəsʔəx%id səshuys tiʔəʔ wəq’əb

bək’ʷ ʔəs–ʔəx%id s=ʔəs–huy=s tiʔəʔ wəq’əb

all STAT–what.happen NM=STAT=be.done=3PO PROX box

‘this box was being done to [i.e., battered about] in everyway’

104

(Bierwert 1996: 183, line 41)

In both of these examples, the expression bək’ʷ ʔəsʔəx&id ‘in every way’ takes a predicate

complement that specifies the event whose circumstances are being referred to. In (341a), this

complement is a bare verb phrase haʔɬ šuɬ ‘be good-looking’ which shares its subject with the

main predicate. In (341b), the complement is a sentential nominalization referring to an event or

set of circumstances pertaining to the subject of the clause as a whole (the box).

104

This line is glossed in the source as ‘this box was doing everything’; however, the context is that the box has

been tossed into the water and is being thrown about by the waves, not that the box was actively doing anything. The

re-glossing provided here compensates for the decontextualization of the sentence, and mirrors the structure of the

Lushootseed sentence at the expense of fluid English.

284

This indefinite anaphoric use of ʔəx&id is particularly frequent in negative expressions like

those in (342):

(342) a. xʷiʔ kʷi stab gʷədsʔəx%id

xʷiʔ kʷi stab gʷə=d=s=ʔəx%id

NEG REM what SBJ=1SG.PO=NM=what.happen

‘there is nothing I could do [to help you]’ (lit. ‘what I could do [is] not [i.e., nothing]’)

[ML Mink and Tutyika I, line 29]

b. xʷiʔ gʷədsʔəx%id

xʷiʔ gʷə=d=s=ʔəx%id

NEG SBJ=1SG.PO=NM=what.happen

‘I didn’t do anything’

[DS Star Child, line 252]

In (342a), ʔəx&id appears as part of a complex subject-phrase, a relative clause headed by stab

‘what?’, itself used as an indefinite anaphor. The existence or reality of the entire subject phrase

(‘that which I could do’) is negated by the clausal predicate, the negative adverb xʷiʔ. In (342b),

ʔəx&id itself is the clausal predicate and is negated by xʷiʔ, which in this case acts as a bona fide

adverb. (For the distinction between to the two types of negative clause, see Section 8.5 below).

Note that in this case ʔəx&id has lost its interrogative sense despite being in predicate position.

In a few other contexts, ʔəx&id is fond used as a clausal predicate with an indefinite rather than

an interrogative meaning. One of these, a narrator’s aside at a point in the story where he is

unable to remember the particular song sung by a character, is given in (343):

(343) ʔu, ləcuʔəx%id

ʔu ləcu–ʔəx%id

INTJ CONT–what.happen

‘or whatever it was’

[DM Basket Ogress, line 55]

More commonly, ʔəx&id used as an indefinite predicate has a temporal sense, as shown in (344):

(344) a. ʔiˑ, xʷuʔələʔ ʔəsʔəx%id tiʔiɬ səsx%əɬs

ʔi xʷuʔələʔ ʔəs–ʔəx%id tiʔiɬ s=ʔəs–x%əɬ=s

INTJ maybe STAT–what.happen DIST NM=STAT–sick=3PO

‘indeed, [Coyote] was sick for quite some time, I guess’

(lit. ‘maybe his sickness [was] happening’)

(Hess 1998: 91, line 26)

285

b. xʷuʔələʔ ʔəsʔəx%id cəlac sləx%il

xʷuʔələʔ ʔəs–ʔəx%id cəlac sləx%il

maybe STAT–what.happen four day

‘maybe [it lasted] four days’

(Bates, Hess & Hilbert 1994: 14)

These examples parallel the temporal use of ʔəx&id shown in (336c). Although the predicative

indefinite uses of ʔəx&id are textually infrequent, it seems likely, given the range of possible

meanings shown by this word in other contexts, that ʔəx&id has an equally wide range of potential

meanings when used as a clausal predicate.

As noted above, ʔəx&id takes the external causative suffix -txʷ (Section 2.1.2.2) to form

ʔəx&itxʷ ‘do what to something?’, and this form in turn takes the dative applicative -yi-d (2.1.3.1),

forming ʔəx&ix&txʷyid ‘do what with something of someone’s?’:

(345) a. ɬələsʔəx%itxʷ kʷədiʔ kʷagʷičəd

ɬu=ləs–ʔəx%id–txʷ kʷədiʔ kʷagʷičəd

IRR=PROG.STAT–what.happen–ECS REM.DMA elk

‘what is he going to do to that elk?’

b. bələcuʔəx%ix%txʷyidəxʷ čəxʷ ti adč’abiqʷ

bə=ləcu–ʔəx%ix%–txʷ–yi–d=əxʷ čəxʷ ti ad–č’abiqʷ

ADD=CONT–what.happen–ECS–DAT–ICS=now 2SG.SUB SPEC 2SG.PO–

great.grandchild

‘what are you doing with [that tallow hoop] of your great-grandson’s?’

(Bates, Hess & Hilbert 1994: 14)

The breakdown of the second of these two forms, ʔəx&ix&txʷyid, is not entirely clear, as it appears

to contain a Type II attenuative reduplication (ʔəx&itxʷyid > ʔəx&ix&itxʷyid) followed by syncope

(ʔəx&ix&itxʷyid > ʔəx&ix&txʷyid). However, as there is only one attestation of the form in the corpus,

it is unknown whether the reduplication in the stem is a productive reduplication specific to this

example, or if this is the fixed form of the stem (perhaps a fossilization of an earlier reduplicated

form).

ʔəx&id is also found combined with another verbal derivational affix, -il ‘inchoative’, in the

form ʔəx&idil ‘happen, come to pass’:

286

(346) ƛ’uʔəx%idil gʷəl lək’ʷit’ tiʔəʔ sʔušəbabdxʷ sbəq’ʷaʔ

ƛ’u=ʔəx%idil gʷəl lə=k’ʷit’ tiʔəʔ sʔušəbabdxʷ sbəq’ʷaʔ

HAB=what.happen–INCH SCONJ PROG=go.waterward PROX poor.guy heron

‘it happened every now and then that poor Heron was going down to shore’

(Hess 2006: 21, lines 229 – 30)

As with the combination of -il and čad, this form seems to lack an interrogative sense, although

its use may be loosely indefinite in the sense that a general type of event is being referred to (or,

more accurately, a set of events repeated over an undefined span of time) instead of a specific

event grounded in a specific time. It may be the case, then, that ʔəx&idil has lost the interrogative

meaning of its root; however, the sentence in (346) is the only attestation in the corpus and so it

is impossible to tell whether there is also an interrogative use open to this form.

105

2.7 Other word classes

2.7.1 Personal and possessive pronouns

Although Lushootseed has a variety of anaphoric devices for referring to and tracking event-

participants and arguments in discourse, the bulk of these seem best treated as parts of the word-

level or phrase-level inflectional systems (see the discussion of object-markers in Section 8.1.2

and subject-markers in 8.1.1 below), or as elements from the deictic system (2.4). There is,

however, a set of anaphoric elements that have the syntactic distribution of nouns and thus lend

themselves to classification as true pronouns. These are given in Table 68:

ʔəca

dibəɬ

dəgʷi

gʷəlapu

cədiɬ

caadiɬ

105

There is also one attestation of another derived form based on ʔəx&id, ʔəxidəb:

(i) bək’ʷ ʔəsʔəx%idəb

bək’ʷ ʔəsʔəx%idəb

all STAT–what.happen

‘it was all different ways’

(Bates, Hess & Hilbert 1994: 14)

However, no other information is available on this use, nor is it apparent from this example what, if any, difference

there is between ʔəx&id and ʔəxidəb. The example is merely provided here in the interests of full disclosure.

287

Table 68: Personal pronouns

Like nouns, these elements appear variously as the arguments of verbs (347a) and nominalized

clauses (b), as the complements of pronouns (c), and as the heads of predicative expressions (d):

(347) a. ʔiˑ, ɬuk’ʷic’yid čəɬ ti dəgʷi

ʔi ɬu=k’ʷic’–yi–d čəɬ ti dəgʷi

INTJ IRR=butcher–DAT–ICS 1PL.SUB SPEC you

‘indeed, we will butcher it for you’

(Hess 1998: 241)

b. hikʷ kʷagʷičəd tiʔiɬ səsɬild čəɬ ti dəgʷi

hikʷ kʷagʷičəd tiʔiɬ s=ʔəs–ɬil–d čəɬ ti dəgʷi

big elk DIST NM=STAT–give.food–ICS 1PL.PO SPEC you

‘our gift of food to you [is] a big elk’

(Hess 1998: 81, line 110)

c. gʷəl xʷiʔ ɬuləx%əb dxʷʔal dəgʷi

gʷəl xʷiʔ ɬu=lə=x%əb dxʷʔal dəgʷi

INTJ not IRR=NEGP=be.heavy for you

‘and it will not be heavy for you’

(Hess 1998: 81, line 92)

d. dəgʷi kʷədaʔ kʷi gʷəƛ’ub gʷəshuyəxʷ ɬukʷaɬ

dəgʷi kʷədaʔ kʷi gʷə=ƛ’ub gʷə=s=huy=əxʷ ɬukʷaɬ

you maybe REM SBJ=well SBJ=NM=be.done=now sun

‘maybe the one who is best to be the sun [is] you’

[HM Star Child, line 172]

Like nouns, pronouns are typically introduced by determiners when heading referential

expressions (Section 7.1) as in (347a) and (b), although this is not always the case, as in (347c).

When the antecedent of the pronoun is female and the determiner is used, the determiner takes its

feminine form:

(

348) ƛʼubəxʷ x%ʷulʼəxʷ ɬadqʼədᶻax% tsiʔiɬ dəgʷi, pʼuayʼ

ƛʼub=əxʷ x%ʷulʼ=əxʷ ɬu=ad–qʼədᶻax tsiʔiɬ dəgʷi pʼuayʼ

good=now just=now IRR=2SG.PO–entrails DIST:FEM you flounder

‘it’s better that they will just become your entrails, Flounder’

[ML Mink and Tutyika, line 184]

288

This example also shows a pronoun in another nominal syntactic role, that of possessor of an NP.

The use of a determiner with a first- or second-person pronoun is extremely unusual from a

typological point of view.

Although the first- and second-person pronouns can function as objects of a predicate as in

(347a), they are rarely the subject of a predication, this role being filled by the various subject-

markers (see Section 8.1 for some discussion). However, there are a few examples in the texts of

a pronoun being used as a subject:

(349) a. ləcuɬalšəbəxʷ dəgʷi, tsi sƛʼalqəb

ləcu–ɬal

–š–əb=əxʷ dəgʷi tsi sƛʼalqəb

PROG.STAT–remove.from.fire–ICS–PASS=now you SPEC:FEM monster

‘you are being taken away out of the fire, Ogress’

[LA Basket Ogress, line 84]

b. hiwil! ʔuxʷ dəgʷi!

hiwil ʔuxʷ dəgʷi

go.on go you

‘go on! you go!’

[HM Star Child, line 170]

c. dəgʷi ʔuyəcəbtubš

dəgʷi ʔu–yəc–əb–txʷ–bš

you PFV–inform–MD–CS–1SG.OBJ

‘you informed me’

(Hess 1967a: 54)

The infrequency of such uses of the pronouns is probably due largely to the requirements of

referent-tracking, which in general requires subjects to be topical and, as such, they are

infrequently realized as overt lexical elements in a clause (Section 11.2.2). While the use of

subject pronouns in imperatives might seem more probable, the example in (349) is the only

attestation of this in the corpus found to date.

In addition to appearing in the roles illustrated in (347), the third-person pronoun also

appears in a textually frequent appositive construction, as in (350):

David Beck 10-2-7 2:19 PM

Comment: note the word order in c!

289

(350) x%əɬ tukʷədadəxʷ tiʔəʔ cədiɬ biac

x%əɬ tu=kʷəda–d=əxʷ tiʔəʔ cədiɬ biac

apparently PAST=grab–ICS=now PROX it meat

‘it seems that he couldn’t get a hold of it, that meat’

(Hess 1998: 85, line 213)

The third-person pronoun is also unusual in that, unlike most other third-person anaphora, it has

a plural form, formed by a Type 6 replication of the first vowel in the stem. The use of the plural

form, however, is not obligatory with plural referents, as illustrated by the following pair of lines

from the story ʔəsɬaɬlil tiʔiɬ ʔi sgʷəlub ʔi tiʔiɬ x&əɬx&əɬtəds ‘Pheasant and his brothers-in-law lived

there’ told by Martha Lamont (Hess 1998):

(

351) a. ƛʼuxʷiʔxʷiʔ tiʔiɬ caadiɬ x%əɬx%əɬtəds

ƛʼu=xʷiʔxʷiʔ tiʔiɬ caadiɬ x%əɬ–x%əɬtəd–s

HAB=hunt DIST they DSTR–brother.in.law–3PO

‘his [Pheasant’s] brothers-in-law would go hunting’

b. gʷəl ƛʼuɬčiˑl tiʔəʔ cədiɬ

gʷəl ƛʼu=ɬčil tiʔəʔ cədiɬ

INTJ HAB=arrive PROX he/she

‘and they would arrive’

(Hess 1998: 263, lines 16-17)

In the first of these two lines, the brothers-in-law of Pheasant are designated as plural both by the

Type 1 reduplication of the nominal root x&əɬtəd ‘brother in law’ and by the choice of the plural

form of the pronoun, caadiɬ. In the subsequent line, however, the same plural referent is

expressed by the singular form of the pronoun, cədiɬ. The optionality of number-marking, even

for human referents, is typical of the Lushootseed treatment of nominal number.

In addition to the personal pronouns listed in Table 68, Lushootseed also has a possessive

pronoun, sgʷaʔ ‘one’s own’, which is inflected using the possessive affixes (Section 7.2) to give

the following paradigm:

dsgʷaʔ

sgʷaʔ čəɬ

adsgʷaʔ

sgʷaʔləp

sgʷaʔs

Table 69: Possessive pronouns

290

Like the other personal pronouns, possessive pronouns can be the subjects (352a), objects

(352b), or complements (352c) of syntactic predicates:

(352) a. gʷəl diɬəxʷ ʔuʔux%ʷtub ʔə tiʔəʔ bəščəb ʔi tiʔəʔ qaw’qs tiʔiɬ sgʷaʔs əlgʷəʔ

gʷəl diɬ=əxʷ ʔu–ʔux%ʷ–txʷ–b ʔə tiʔəʔ bəščəb ʔi tiʔəʔ qaw’qs

SCONJ FOC=now PFV–go–ECS–PASS PR PROX mink and PROX raven

tiʔiɬ sgʷaʔ–s əlgʷəʔ

DIST one’s.own–3PO PL

‘and theirs (i.e., their share) was taken by Mink and Raven’

(Hess 1998: 73, line 199)

b. luləskʷədad čəxʷ ti diɬ dsgʷaʔ dsʔabyicid

lu=ləs–kʷəda–d čəxʷ ti diɬ d–sgʷaʔ

IRR=PROG.STAT–taken–ICS 2SG.SUB SPEC FOC 1SG.PO–one’s.own

d=s=ʔab–yi–t–sid

1SG.PO=NM=extend–DAT–ICS–2SG.OBJ

‘you will be taking this very one of mine which I give to you’

(Hess 2006: 73, line 733)

c. xʷiʔ lədsgʷaʔ

xʷiʔ lə=d–sgʷaʔ

NEG NEGP=1SG.PO–one’s.own

‘they are not mine’

(Hess 1998: 85, line 200)

However, like the personal pronouns, the possessive pronouns are textually far more frequent as

syntactic predicates:

(353) a. xʷuʔələʔ sgʷaʔləp tiʔiɬ sʔələd, tatačulbixʷ

xʷuʔələʔ sgʷaʔ–ləp tiʔiɬ sʔələd tatačulbixʷ

maybe one’s.own–2PL.PO DIST food big.game.animal

‘maybe that food, big game animal, is yours’

(Hess 1998: 80, line 62)

b. adsgʷaʔ ʔu tiʔiɬ

ad–sgʷaʔ ʔu tiʔiɬ

2SG.PO–one’s.own INT DIST

‘is that yours?’

(Hess 1998: 85, line 204)

291

c. adsgʷaʔ adshuyalc tiʔəʔ cəxʷhuyəxʷ ʔəsʔistaʔ

ad–sgʷaʔ ad=s=huy–alc tiʔəʔ d=dəxʷ=huy=əxʷ

2SG.PO–one’s.own 2SG.PO=NM=made–PRDCT PROX 1SG.PO=ADNM=made=now

ʔəs–ʔistaʔ

STAT–be.like

‘it is your doing, that I have been made to be like this’

[ML Mink and Tutyika II, line 91]

As seen in (353c), the possessive pronouns can potentially take bare nominal complements,

indicating the type of possessed item the pronoun refers back to:

(354) a. bəsgʷaʔsəxʷ bəsxʷiʔxʷiʔs kʷədiʔ sucutcuts

bə=sgʷaʔ–s=əxʷ bə=sxʷiʔxʷiʔ–s kʷədiʔ s=ʔu–cut–cut=s

ADD=one’s.own–3PO=now ADD=game=3PO REM.DMA NM=PFV–DSTR–say=3PO

‘it was his own game, he was saying’

(Hess 1998: 87, line 242)

b. xʷiʔ lədsgʷaʔ dsqʷəbayʔ

xʷiʔ lə=d–sgʷaʔ d–sqʷəbayʔ

NEG NEGP=1SG.PO–one’s.own 1SG.PO–dog

‘they are not my dogs’

(Hess 1998: 79, line 33)

c. diɬəxʷ ɬusgʷaʔs ɬusəsyayus ʔə tiʔəʔ təkʷtəkʷəlus

diɬ=əxʷ ɬu=sgʷaʔ–s ɬu=s=ʔəs–yayus ʔə tiʔəʔ təkʷtəkʷəlus

FOC=now IRR=ones.own=3PO IRR=NM=STAT–work PR PROX owl

‘it will be Owl’s very own work’

(Hess 2006: 9, line 162)

Like the possessive pronoun itself, the complement is inflected for the person and number of its

possessor. The complement can be a concrete noun like sxʷiʔxʷiʔ ‘game’ as in (354a) or

sqʷəbayʔ ‘dog’ in (354b), or it can be an s=nominal representing a more abstract concept, as in

(354c). Like other type of bare-nominal complements, these nouns appear without determiners

and rigidly follow the head of the NP.

292

2.7.2 Quantifiers

2.7.3 Numerals

Lushootseed has three series of words used for counting. The first of these are the general,

plain-series numerals used for counting non-humans:

dəč’uʔ

x&ʷəlačiʔ

saliʔ

100

dəč’uʔ sbək’ʷačiʔ

ɬixʷ

101

dəč’uʔ sbək’ʷačiʔ ʔi kʷi dč’uʔ

buus

102

dəč’uʔ sbək’ʷačiʔ ʔi kʷi saliʔ

cəlac

110

dəč’uʔ sbək’ʷačiʔ ʔi kʷi ʔulub (NL)

yəlaʔc (NL), dᶻəlačiʔ (SL)

111

dəč’uʔ sbək’ʷačiʔ ʔi kʷi ʔulub ʔi kʷi dč’uʔ (NL)

c’ukʷs

120

dəč’uʔ sbək’ʷačiʔ ʔi kʷi saliʔačiʔ

təqačiʔ

200

saliʔ sbək’ʷačiʔ

x&ʷəl

300

ɬixʷ sbək’ʷačiʔ

ʔulub (NL), padac (SL)

400

buus sbək’ʷačiʔ

ʔulub ʔi kʷi dč’uʔ (NL)

106

500

cəlac sbək’ʷačiʔ

ʔulub ʔi kʷi saliʔ (NL)

600

yəlaʔc sbək’ʷačiʔ (NL), dᶻəlačiʔ sbək’ʷačiʔ (SL)

saliʔačiʔ

700

c’ukʷs sbək’ʷačiʔ

saliʔačiʔ ʔi kʷi dč’uʔ

800

təqačiʔ sbək’ʷačiʔ

saliʔačiʔ ʔi kʷi saliʔ

900

x&&əl sbək’ʷačiʔ

sɬixʷačiʔ

1000

ʔulub sbək’ʷačiʔ (NL), padac sbək’ʷačiʔ (SL)

sbuusačiʔ

1001

ʔulub sbək’ʷčiʔ ʔi kʷi dč’uʔ (NL)

scəlacačiʔ

1100

ʔulub ʔi kʷi dč’uʔ sbək’ʷačiʔ (NL)

107

yəlaʔcačiʔ (NL), dᶻəlačiʔačiʔ (SL)

1900

ʔulub ʔi kʷi x&ʷəl sbək’ʷačiʔ (NL)

c’ukʷsačiʔ

2000

saliʔačiʔ sbək’ʷačiʔ

təqačiʔačiʔ

3000

sɬixʷačiʔ sbək’ʷačiʔ

Table 70: Cardinal numerals—plain series

Several of the forms in Table 70 are analyzable as consisting of a radical plus the lexical suffix

-ačiʔ ‘hand’. In both Northern and Southern Lushootseed, this suffix is used to indicate decades

(that is, the interval corresponding to the number of fingers on both hands), while the word for

100, sbək’ʷačiʔ, appears to contain the radical √bək’ʷ ‘all’, suggesting an etymological source in

an expression meaning ‘all hands’. The word for eight, təqačiʔ, is based on the radical √tq

‘closed’, ‘closed hands’ perhaps being a reference to the number indicated by holding up both

106

Literally, ‘ten and one (thing)’. In Southern Lushootseed, the form for eleven is padac yəxʷ kʷi dəč’uʔ (Hess &

Hilbert 1976), which would have the same literal gloss.

107

Tweddell (1950) reports 1,100 (given here in standardized transcription) as padac sbək’ʷačiʔ yəxʷ kʷi dč’uʔ

sbək’ʷačiʔ; however, this form does not follow the pattern for the other thousand-hundred forms found in the same

work and in Hess & Hilbert (1976).

293

hands with the thumbs folded over the palms.

108

The Southern Lushootseed word for six,

dᶻəlačiʔ, is based on º√dᶻəl ‘turned, changed’, which probably derives from the fact that, when

counting on one’s fingers, six requires the use of the fingers on the next hand. The words for 30,

40, 50, and 100 also begin with the nominalizing prefix s- (Section 2.2.1); this prefix is

occasionally found with the decades from 60 to 90 as well.

Numbers that combine decades or centuries with digits such as the Northern Lushootseed

words for 11 (ʔulub ʔi kʷi dč’uʔ), 21 (saliʔačiʔ ʔi kʷi dč’uʔ), or 101 (dəč’uʔ sbək’ʷačiʔ ʔi kʷi

dč’uʔ) use the conjunction ʔi ‘and’ (yəxʷ is used in the Southern Lushootseed equivalents) and a

determiner to connect the digits to the remainder of the expression. The choice of determiner is

governed by the same considerations of specificity, uniqueness, and deixis that govern

determiner choice in other contexts (see Section 2.4 above), kʷi being used as a default when one

is counting in the abstract and not enumerating concrete objects. Orders of centuries are specified

by combining words for digits and decades with sbək’ʷačiʔ ‘100’, making the Lushootseed

numerals between 100 and 10,000 the equivalent of those used in spoken English for numbers

(e.g., cəlac sbək’ʷačiʔ = five hundred, ʔulub ʔi kʷi x&ʷəl sbək’ʷačiʔ = nineteen hundred),

although in Lushootseed — but not in English — this pattern is extended to the even millennia

(ʔulub sbək’ʷačiʔ ‘one thousand’ [lit. ‘ten hundred’]).

109

Lushootseed has a special series of numerals used specifically for counting people, given in

Table 71:

diič’uʔ

t$qqačiʔ

səsaʔliʔ

x,ʷ$l$l

ɬixʷixʷ

ʔululub (NL), padadac (SL)

bəbuʔs

ʔululub ʔi diič’uʔ (NL)

110

c$l$lac

salaʔačiʔiʔ (SL)

108

Hess & Hilbert (1976: 28) suggest the opposite pattern, counting down with the fingers until all are folded over

the palm except the thumbs.

109

Tweddell (1950: 72) also reports the borrowing ta’wowsəd [sic] ‘thousand’.

110

The form padadačiʔ yəxʷ kʷi diič’uʔ is given for Snoqualamie-Duwamish by Tweddell (1950: 72), who

reports the form padadačiʔ ‘ten’ as being used in compound numbers from 11–19. The form padadac is used for

10 and as the initial member of compounds based thereon — e.g., padadac bəkʷ’ačiʔ ‘1000’.

294

yələlaʔc (NL), dᶻəlalačiʔ (SL)

100

diič’uʔ bəkʷ’ačiʔ (SL)

c’ukʷukʷs

1000

padadac bəkʷ’ačiʔ (SL)

Table 71: Cardinal numerals — human series

With the exception of the words for one, two, and four people, this series is formed from the

plain-series numerals by Type III reduplication (Section 5.3.4). For numerals between 10 and 20

that combine the first decade with a digit, both numerals are taken from the human series:

(355) ʔululub ʔi diič’uʔ

ʔululub ʔi diič’uʔ

ten:HMN and one:HMN

‘eleven people’

[LA Basket Ogress, line 56]

In Northern Lushootseed, the human series of numerals is only used up to 20, after which plain-

series numerals are used. In Southern Lushootseed, the complex numerals for 100 and 1000

people are given in Tweddell (1950: 72) with the initial numerals in the corresponding human

form, diič’uʔ bəkʷ’ačiʔ ‘100 people’ and padadac bəkʷ’ačiʔ ‘1,000 people’ (lit. ‘ten hundred

people’); the higher thousands such as 2,000 and 5,000 use only the plain cardinal forms (i.e.,

saliʔačiʔ sbək’ʷačiʔ ‘2,000 people’ [lit. ‘twenty-hundred people’] and cəlacačiʔ sbək’ʷačiʔ

‘5,000 people’ [lit. ‘fifty-hundred people’]).

A third series of numerals is used for counting time or iteration:

111

dəč’axʷ

sɬixʷačiʔaɬ

cəbab

sbuusačiʔaɬ

ɬixʷaɬ

scəlacačiʔaɬ

buusaɬ

dᶻəlačiʔačiʔaɬ (SL)

cəlacaɬ

c’ukʷsačiʔaɬ

yəlaʔcaɬ (NL), dᶻəlačiʔaɬ (SL)

təqačiʔačiʔaɬ

c’ukʷaɬ

x&ʷəlačiʔaɬ

təqačiʔaɬ

100

dəč’uʔ sbək’ʷačiʔaɬ

x&ʷəɬaɬ

101

dəč’uʔ sbək’ʷačiʔaɬ yəxʷ kʷi dč’uʔ

ʔulubaɬ (NL), padacaɬ (SL)

200

saliʔ sbək’ʷačiʔaɬ

ʔulubaɬ ʔi kʷi dč’uʔ (NL)

700

cukʷs sbək’ʷačiʔaɬ

ʔulubaɬ ʔi kʷi saliʔ (NL)

1000

padac sbək’ʷačiʔaɬ (SL)

saliʔačiʔaɬ

1001

padac sbək’ʷčiʔaɬ yəxʷ kʷi dč’uʔ (SL)

saliʔačiʔaɬ ʔi kʷi dč’uʔ

1100

padac yəxʷ kʷi dč’uʔ sbək’ʷačiʔaɬ (SL)

saliʔačiʔaɬ ʔi kʷi saliʔ

2000

saliʔačiʔ sbək’ʷačiʔ

Table 72: Cardinal numerals—temporal-iterative series

111

The forms higher than 20 in this table are Snoqualmie-Duwamish forms taken from Tweddell (1950: 72) and

given here in standardized transcription.

295

The first of these forms, dəč’axʷ ‘once’, is commonly truncated to č’əʔaxʷ. The second form

cəbab ‘twice’ is based on a bound suppletive form of the radical. It usually appears in

combination with certain lexical suffixes and varies with the independent form of the radical,

saliʔ. Other than these two, the remainder of the temporal-iterative numerals are formed by

combining the plain-series numeral with the lexical suffix -aɬ ‘times’. For simple numerals, this

suffix is simply added to the stem; for complex numerals, it appears as a suffix on the decade or

century rather than on the digits. The same lexical suffix is also used to form the expressions

qahaɬ ‘many (qa) times’ and dᶻixʷaɬ ‘first (dᶻixʷ) time’.

The temporal-iterative numeral series is also the basis of a few complex temporal expressions

derived using the inchoative suffix -il to create verbs indicating the completion of an action a

specified number of times:

(356) a. x%ʷul’ ʔubuusaɬil tiʔiɬ shuyuds

x%ʷul’ ʔu–buus•aɬ–il tiʔiɬ s=huyu–d=s

only PFV–four•times–INCH DIST NM=be.done–ICS=3PO

‘she had just done it four times’

[MW Star Child, line 130]

b. ʔal kʷi sɬixʷaɬils, huy, kʷədabacdubutəxʷ tiʔəʔ ʔəskikəwič

ʔal kʷi s=ɬixʷ•aɬ–il=s huy kʷəd•abac–dxʷ–but=əxʷ

PR REM NM=three•times–INCH=3PO SCONJ take•body–DC–REFL=now

tiʔəʔ ʔəs–ki–kəwič

DIST STAT–ATTN–hunchbacked

‘on the third time, well, the little hunchbacked one caught his body [on the snags]’

[AJ Basket Ogress, line 57]

It may be, however, that this particular type of derivation is not limited to temporal-iterative

numerals as there are two examples, given in (357), of inchoative verbs formed with a numeral

and a different lexical suffix, -gʷiɬ ‘canoe’:

(357) a. ləcəbagʷiɬil tiʔəʔ ʔəsx%ək’ʷtxʷ əlgʷəʔ ƛ’əlayʔ

lə=c$b•a•gʷiɬ–il tiʔəʔ ʔəs–x%ək’ʷ–txʷ əlgʷəʔ ƛ’əlayʔ

PROG=two•CNN•canoe–INCH PROX STAT–overturned–ECS PL canoe

‘it was coming to be that they had two canoes overturned’

296

b. ləɬixʷalgʷiɬil tiʔəʔ tasx%ək’ʷtxʷ həlgʷəʔ ƛ’əlayʔ dəxʷəsax%ʷəbabac

lə=ɬixʷ•al•gʷiɬ–il tiʔəʔ tu=ʔas–x%ək’ʷ–txʷ həlgʷəʔ ƛ’əlayʔ

PROG=three•CNN•canoe–INCH PROX PAST=STAT–overturned–ECS PL canoe

dəxʷ=lə=sax%ʷəb•abac

ADNM=PROG=jump•body

‘it was coming to be they had three canoes overturned that they were jumping over’

(Hilbert & Hess 1977: 15)

This suggests that the numeral + lexical suffix + inchoative expression is more productive than

its frequency in the corpus indicates, but this will have to remain an open question until

subsequent textual analysis brings more examples to light.

Beyond this, numerals in all three series are surprisingly inert in morphological terms. Other

than expressions of the type shown in (356), numerals seem only to appear as stems in two

related sets of compound words formed with incorporative suffix -aɬ and the radical √dat ‘day’:

cəbdat ‘two days, second day’

scəbdatil ‘Tuesday’

sɬixʷəɬdat ‘three days, third day’

112

sɬixʷəɬdatil ‘Wednesday’

buusəɬdat ‘four days, fourth day’

buusəɬdatil ‘Thursday’ (NL)

—

scəlacəɬdatil ‘Friday’ (NL)

113

Table 73: Compound numeral expressions for days and days of the week

These seem likely to be recent formations introduced along with the European calendar. The fact

that there are a maximum of four forms in each set follows from the naming practices for the

days the week, the remainder of which are č’itabac ‘Saturday’ (from √č’it ‘near’ + -abac

‘body’), x&aʔx&aʔaɬdat ‘Sunday’ (√x&aʔx&aʔ ‘sacred, taboo’), and bəlx&ʷəɬdat ‘Monday’ (√bəlx&ʷ

‘pass by, come after’). Furthermore, it seems possible that the words in the lefthand column of

Table 73 are backformations from the (more morphologically complex) words in the righthand

column, given the absence of potential forms for expressing periods of time in days that do not

correspond to the Lushootseed names for days of the week that happen to be based on numerals.

It should be noted, however, that there are two verbs — ɬixʷəɬdacut ‘become three days’ and

buusaɬdalicut ‘become four days’ (each attested only once in the corpus) — which appear to be

112

This form is also glossed as ‘Wednesday’ in the Lushootseed Dictionary (p. 147).

113

The Lushootseed Dictionary records this form as scəlacəɬdat(il) (p. 45), the parenthesis indicating that the word

is also attested without the inchoative suffix but with the same gloss.

297

derived from sɬixʷəɬdat ‘three days, third day’ and buusəɬdat ‘four days, fourth day’,

respectively. These two verbs are formed through a rather non-transparent use of reflexive

morphology (-cut consisting of the event-internal causative -t and the reflexive marker -sut), and

it remains to be seen whether they are representative of a more extensive set of older forms or if

they are recent coinages of some kind.

All three series of numerals are used both as nominal quantifiers and as clausal predicates.

The former use is illustrated in (358):

(358) a. hay gʷəl, tusulayitəbəxʷ ʔə tiʔiɬ saliʔ sq’ʷiq’ʷəlaƛ’əd

hay gʷəl tu=sula–yi–t–əb=əxʷ ʔə tiʔiɬ saliʔ sq’ʷi–q’ʷəlaƛ’əd

SCONJ SCONJ PAST=centred–DAT–ICS–PASS=now PR DIST two ATTN–berry

‘and then he set before him two little berries’

(Hess 1995: 152, line 21)

b. gʷəl ʔahəxʷ tiʔəʔ səsaliʔ sɬəɬadəyʔ ləqaladibid

gʷəl ʔa=həxʷ tiʔəʔ səsaliʔ sɬə–ɬadəyʔ ləq•al•adiʔ–bi–d

SCONJ be.there=now PROX two:HMN PL–woman hear•CNN•ear–MAP–ICS

‘and there are two women who overhear it’

[HM Star Child, line 26]

c. xʷuʔələʔ ʔəsʔəx%id cəlac sləx%il kʷədiʔ tusaxʷəbtubs

xʷuʔələʔ ʔəsʔəx%id cəlac sləx%il kʷədiʔ tu=s=saxʷəb–txʷ–b=s

maybe STAT–what.happen five day REM.DMA PAST=NM=run–ECS–PASS=3PO

‘maybe it was five days (since) they had been run off with (i.e., kidnapped)’

(Hess 2006: 53, line 264)

Numerals are most frequently used with the singular form of nouns (358a and c), although the

plural form is an option, particularly when referring to numbers of people (358b). As shown in

these examples, numerals may be used in nominal argument phrases (358a and b) as well as in

nominal predicate complements (358c).

As in many languages, the numeral ‘one’, dəč’uʔ, can be used to convey a sense of

specificity or particularity:

298

(359) a. gʷəl huy bəʔəy’ədxʷəxʷ tiʔəʔ qa tiʔiɬ sʔuladxʷ ʔal tiʔiɬ cədiɬ dəč’uʔ stuləkʷ

gʷəl huy bəʔəy’ədxʷəxʷ tiʔəʔ qa tiʔiɬ sʔuladxʷ ʔal tiʔiɬ

SCONJ SCONJ ADD=found–DC=now PROX many DIST salmon at DIST

cədiɬ dəč’uʔ stuləkʷ

he one river

‘and then he found a lot of salmon in this one river’

(Hess 1998: 66, line 23)

b. gʷəl diɬ xʷuʔələʔ tušac’s tiʔiɬ dəč’uʔ syəyəhub

gʷəl diɬ xʷuʔələʔ tu=s=šac’=s tiʔiɬ dəč’uʔ syəyəhub

SCONJ FOC maybe PAST=NM=end=3PO PROX one story

‘and I guess that is the end of this one traditional story’

(Hess 1998: 76, line 279)

c. ʔaləxʷ tiʔəʔ sʔahilsəxʷ tiʔəʔ sɬčils dxʷʔal tiʔəʔ dəč’uʔ swətixʷtəd

ʔal=əxʷ tiʔəʔ s=ʔah–il=s=əxʷ tiʔəʔ s=ɬč–il=s

at=now PROX NM=be.there–INCH=3PO=now PROX NM=arrive–INCH=3PO

dxʷ–ʔal tiʔəʔ dəč’uʔ swətixʷtəd

CNTRPT–at PROX one tree

‘when (she) got to this one tree’

[MS Basket Ogress, line 38]

d. gʷəl ƛ’ubəkʷədad tiʔiɬ bədəč’uʔ qʷɬayʔ

gʷəl ƛ’u=bə=kʷəda–d tiʔiɬ bə=dəč’uʔ qʷɬayʔ

SCONJ HAB=ADD=taken–ECS DIST ADD=one stick

‘then she would take another one of her scratching sticks’

[AW Basket Ogress, line 118]

This use of the numeral is not textually infrequent, and finds a very close parallel in its literal

English gloss.

Like other adnominal modifiers, numerals have some flexibility as to whether they precede

the noun they modify (as in 358) or follow it, as in (360):

(360) a. təɬ čəd ʔuʔəy’dxʷ tiʔiɬ dsqa cəxʷsqatəd səsaʔliʔ ʔal tudiʔ

təɬ čəd ʔu–ʔəy’–dxʷ tiʔiɬ d–sqa

truly 1SG.SUB PFV–found–DC DIST 1SG.PO–older.sibling

d=dəxʷ=sqatəd səsaʔliʔ ʔal tudiʔ

1SG.PO=ADNM=older.siblings two:HMN at DIST.DMA

‘I truly did find my older brothers, my two older brothers over there’

(Hess 2006: 73, line 736)

299

b. sx%aʔhus tsiʔəʔ čəgʷas diič’uʔ

sx%aʔhus tsiʔəʔ čəgʷas diič’uʔ

sawbill PROX:FEM wife one:HMN

‘one wife [was] Sawbill’

(Hess 2006: 22, line 5)

Post-nominal ordering for numerals is rare (these are the only two examples in the current

corpus) and it is not clear what conditions may apply to this ordering.

Also like other modifiers, numerals can be combined with additional adnominal elements in a

single noun phrase:

(361) a. saʔ saliʔ sq’ʷəlaɬəd

saʔ saliʔ sq’ʷəlaɬəd

bad two berry

‘[there were] two measly berries’

(Hess 1995: 152, line 26)

b. dəč’uʔ haʔɬ syəyəhub

dəč’uʔ haʔɬ syəyəhub

one good story

‘[it is] a good story’

(Hilbert & Hess 1977: 32)

The relative ordering of the numeral with respect to other adnominal modifiers appears to be

relatively free, determined by considerations of style or communicative structure.

Numerals are often found as the heads of anaphoric nominal expressions:

(362) a. kʷədad tiʔəʔ dəč’uʔ

kʷəda–d tiʔəʔ dəč’uʔ

taken–ICS PROX one

‘he took one (berry)’

(Hess 1995: 152, line 29)

b. gʷəl bəʔəsgʷədil tsiʔiɬ dəč’uʔ ʔal kʷi xʷqʷəq’ʷus

gʷəl bə=ʔəs–gʷəd–il tsiʔiɬ dəč’uʔ ʔal kʷi xʷqʷəq’ʷus

SCONJ ADD=STAT–down–INCH DIST:FEM one at REM cliff

‘and one (sister) was sitting on the cliff’

[JS Basket Ogress, line 74]

These expressions only occur in contexts where the identity of the item(s) being counted is

recoverable from discourse. Numerals also head NPs in constructions such as that in (363):

300

(363) gʷəl ʔəbsqʷəbqʷəbayʔ ʔə tiʔəʔ bəsaliʔ

gʷəl ʔəs–bəs–sqʷəb–qʷəbayʔ ʔə tiʔəʔ bə=saliʔ

SCONJ STAT–PROP–DSTR–dog PR PROX ADD=two

‘and (they) had two dogs, too’

(Hess 1998: 78, line 23)

This is a usual way of expressing the notion ‘X has a certain number of Y’ when the fact of

possession, rather than the number possessed, is the focus of the utterance (cf. the sentences

in 368 below).

Perhaps even more frequently that they are found inside nominal argument expressions,

numerals occur in predicate position as the heads of clauses:

(364) saliʔ kʷi ɬuʔəƛ’txʷ čəxʷ č’ƛ’aʔ

saliʔ kʷi ɬu=ʔəƛ’–txʷ čəxʷ č’ƛ’aʔ

two REM IRR=come–ECS 2SG.SUB stone

‘the stones that you will bring [will be] two’

[AW Basket Ogress, line 80]

b. saliʔ tiʔiɬ ʔəskʷədad

saliʔ tiʔiɬ ʔəs–kʷəda–d

two DIST STAT–taken–ICS

‘what he is holding [are] two (halibut)’

(Hess 2006: 55, 328)

c. ɬixʷixʷ tiʔəʔ caadiɬ təlixʷ suq’ʷaʔ

ɬixʷixʷ tiʔəʔ caadiɬ təlixʷ suq’ʷaʔ

three:HMN PROX they blood.brother younger.sibling

‘these full-blood brothers [were] three’

(Hess 1998: 73, line 204)

Like other non-verbal predicates, numerals in this context regularly take clitics for mood and

tense:

(365) gʷəl gʷətusəsaliʔ əlgʷəʔ

gʷəl gʷə=tu=səsaliʔ əlgʷəʔ

SCONJ SBJ=PAST=two:HMN PL

‘and there would have been two of them’

(Hess 2006: 17, line 136)

There is also one example in the corpus of a numeral taking an aspectual prefix:

301

(366) ʔəsbuus kʷi tuhuyud əlgʷəʔ

ʔəs–buus kʷi tu=huyu–d əlgʷəʔ

STAT–four REM PAST=make–ICS PL

‘what they made [was] four [moccasins]’

[DS Star Child, line 355]

However, the fact that there is a single occurrence of an aspect-marker on a numeral in the

corpus suggests that this is a rather marginal construction. That it occurs on the numeral buus

‘four’ may not be coincidental, given that four is a culturally important numeral in Northern

Lushootseed (Hess 1995),

114

suggesting perhaps that buus might be a prime candidate for

recategorization as a verb to express the notion of making/doing something in fours.

Like any other predicate, numerals can head yes/no interrogatives and can take ordinary

matrix-clause subject markers:

(367) a. səsaliʔ ʔu

səsaliʔ ʔu

two:HMN INT

‘were there two (children)?’

(Hess 2006: 43, line 37)

b. ɬixʷačiʔəxʷ čəd ʔi kʷi yəlaʔc

ɬixʷačiʔ=əxʷ čəd ʔi kʷi yəlaʔc

thirty=now 1SG.SUB and REM six

‘I’m thirty-six now’

(Bates, Hess & Hilbert 1994: 277)

As shown in (367b), complex numerals in predicate position are treated like other multi-word

predicate phrase in terms of the placement of subject-markers and other sentence-second clitics.

Numerals in both argument and predicate phrases are frequently combined with lexical

suffixes (Section 2.8) acting as numeral classifiers:

(368) a. … tusaxʷəbabacəd tiʔəʔ buusalgʷiɬ ƛ’əlayʔ

tu=saxʷəb•abac–əd tiʔəʔ buus•al•gʷiɬ ƛ’əlayʔ

PAST=run•body–ICS PROX four•CNN•canoe shovel.nose.canoe

‘… [they] jumped over the four shovel-nosed canoes’

(Hilbert & Hess 1977: 16)

114

The number 4 is also culturally important in a number of neighbouring but unrelated languages, and the form

*moos proposed by Edward Sapir is a plausible early borrowing across languages of the area (Lushootseed having

changed *m to /b/ in the mid-Nineteenth Century). Among the Southern Lushootseed, the numeral is 5.

302

b. cəbagʷiɬəxʷ kʷi ɬudəxʷsaxʷəbabacəds əlgʷəʔ

cəb•a•gʷiɬ=əxʷ kʷi ɬu=dəxʷ=saxʷəb•abac–əd=s əlgʷəʔ

two•CNN•canoe=now REM IRR=ADNM=jump•body–ics=3PO PL

‘now what they were jumping over [was] two canoes’

(Hilbert & Hess 1977: 15)

c. č’əʔilc ti dtalə

č’əʔ•ilc ti d–talə

one•round SPEC 1SG.PO–dollar

‘I have one dollar’

(Hess & Hilbert 1976: I, 68)

d. ʔəsɬixʷulč ti dsʔax%ʷuʔ

ʔəs–ɬixʷ•ulč ti d–sʔax%ʷuʔ

STAT–three•container SPEC 1SG.PO–clam

‘I have three clams’

(Hess 1995: 20)

The choice of suffix is largely semantic: lexical suffixes with very concrete, specific meanings

are used in counting those objects which they designate (as in 368a and b), while others are used

when counting objects that fit into the general class (in either shape or function) of object

expressed by the suffix (as in 368c and d). The use of lexical suffixes as classifiers is discussed

in more detail in Section 4.3 below. Also illustrated in (368b) and (c) are the alternate combining

forms for the numerals one and two, č’əʔ- and ‘one’ cəb- ‘two’, noted above in the context of the

temporal-iterative series of numerals in Table 72. These forms seem to be in free variation with

the regular stems, dəč’uʔ ‘one’ and saliʔ ‘two’ in enumerative constructions.

2.7.4 Directional particles

There are three directional particles in Lushootseed that are used to express direction of

movement or point of origin — centripetal dxʷ, centrifugal tul’, and prolative liɬ. These particles

may be used to modify verbal predicates and nouns phrases, as shown in the following example:

303

(369) gʷəč’axʷadəxʷ čəxʷ ʔə kʷi tul’ tiʔəʔ haʔɬ buʔqʷ ʔə tiʔəʔ dx%ʷubt

gʷə=č’axʷa–d=əxʷ čəxʷ ʔə kʷi tul’ tiʔəʔ haʔɬ buʔqʷ ʔə

SBJ=clubbed–ECS=now 2SG.SUB PR REM CNTRFG PROX good duck PR

tiʔəʔ d–x%ʷubt

PROX 1SG.PO–paddle

‘you could club them from [among] the good waterfowl with my paddle’

(Hess 2006: 62, line 485)

Although the sense of tul’ is difficult to capture in the English gloss, it is used here to convey the

sense of the action originating from a point within the flock of waterfowl who, in the story, are

attacking the people hosting the protagonists. Hess (p.c.) reports that uses of the directional

particles as modifiers of nouns and verbs is not uncommon in ordinary speech; however, (369) is

the only such example in the textual corpus used for this study. In Southern Lushootseed, tul’ is

sued to introduce the standard of comparison in comparative constructions (Section 8.8):

(370) ʔəswələx%ʷ čəxʷ tul’ ʔəca

ʔəs–wələx%ʷ čəxʷ tul’ ʔəca

STAT–strong 2SG.SUB CNTRFG I

‘you are stronger than I’

(Hess & Hilbert 1976: II, 49)

The Northern Lushootseed counterpart to this is the preposition dxʷʔal.

More commonly than they are used as independent lexical items, the directional particles are

found compounded with words belonging to other lexical classes. In these uses, the directional

particles are most frequent in the derivation of the various locative prepositions (Section 2.3.1),

but they are also found associated with a few verbs expressing location (e.g., √ʔa ‘be there’, √č’it

‘nearby’, √lil ‘far’), locative adverbs (Section 2.5.2), and interrogative words (8.4.2 below). As

with other quasi-inflectional morphology, directional particles do not alter the part of speech of

the radical to which they are attached. Thus, interrogatives that take directional particles remain

interrogatives and adverbs that take directionals remain locative adverbs, as in (371):

304

(371) a. tul’čad kʷi skʷədxʷs əlgʷəʔ

tul’–čad kʷi s=kʷəd–dxʷ=s əlgʷəʔ

CNTRFG–where REM NM=taken–DC=3PO PL

‘from where did they manage to get it?’

(Hess 1998: 83, line 160)

b. tuxʷit’ilcut tul’šəq

tu=xʷit’il–t–sut tul’–šəq

PAST=come.down–ICS–REFL CNTRFG–high

‘she came down from above’

[DS Star Child, line 5]

Directional particles affixed to an interrogative or adverb appear immediately adjacent to the

radical, inside the proclitic string.

(372) a. ʔa kʷi ƛ’ubəsliɬt’aq’ts

ʔa kʷi ƛ’u=bə=s=liɬ–t’aq’t=s

be.there REM HAB=ADD=NM=PRLV–inland=3PO

‘there he would be up along the bank [i.e., just inland of the water]’

(Hess 1998: 67, line 54)

b. stab kʷi gʷəcəxʷtəɬəɬ dxʷgʷəd dxʷʔal ti tucəxʷtul’ʔa

stab kʷi gʷə=d=dəxʷ=təɬəɬ dxʷ–gʷəd dxʷ–ʔal ti

what REM SBJ=1SG.PO=ADNM=arrive.safely CNTRPT–down CNTRPT–at SPEC

tu=d=dəxʷ=tul’–ʔa

PAST=1SG.PO=ADNM=CNTRFG–be.there

‘what way can I get down safely to where I am from?’

(Hess 2006: 28, line 144)

Because locative adverbs do not take aspectual or other types of prefix, directionals are not

ordered relative to other prefixal morphemes.

The particle dxʷ ‘centripetal [CNTRPT]’ expresses motion towards a region or location:

115

(373) a. ɬuʔibəš čəd čəda ɬuʔux%ʷ dxʷt’aq’t

ɬu=ʔibəš čəd čəda ɬu=ʔux%ʷ dxʷ–t’aq’t

IRR=travel 1SG.SUB 1SG.COORD IRR=go CNTRPT–inland

‘I will travel and I will go up there inland’

(Hess 1998: 78, line 15)

115

Although the particle dxʷ ‘CENTRIPETAL’ is generally homophonous with the prefix dxʷ- ‘contained’, the latter

has a voiceless alternate [txʷ] in rapid speech and may be realized as [dəxʷ] in casual but less rapid speech, while

the former remains [dxʷ] in both contexts. Etymologically, the two prefixes clearly have different origins (Hess

1971).

305

b. gʷəl ʔix%ʷitəb dxʷčaʔkʷ dxʷʔal qʷuʔ

gʷəl ʔix%ʷi–t–əb dxʷ–čaʔkʷ dxʷ–ʔal qʷuʔ

SCONJ thrown.away–ICS CNTRPT–seaward CNTRPT–at water

‘and he threw him towards the sea [and] into the water’

(Hess 1998: 69, line 125)

c. siʔ ƛ’ub ɬadsəsƛ’ip’aɬəd čx%a ɬuɬčil dxʷdiʔaʔ …

siʔ ƛ’ub ɬu=ad=s=ʔəs–ƛ’ip’–aɬ–əd čx%a ɬu=ɬčil

SCONJ well IRR=2SG.PO=NM=STAT–encircled–INCRP–ICS 2SG.CORD IRR=arrive

dxʷ–diʔaʔ

CNTRPT–PROX.DMA

‘just so will you be clutching it firmly and you will arrive here’

(Hilbert & Hess 1977: 19)

In the last sentence here, (373c), the directional particle appears not on a locative adverb but on a

demonstrative adverbial, diʔaʔ ‘PROXIMAL’. As this is the only such example in the present

corpus, it is not clear if dxʷdiʔaʔ is a lexicalized form, or if the directionals are generally

applicable to demonstrative adverbials.

In addition to appearing as adverbials as in the examples in (373), forms with dxʷ are also

potentially clausal predicates, as in (374):

(374) a. tiˑləb dxʷt’aq’t tiʔiɬ sɬalil ʔə tiʔiɬ čxʷəluʔ

tiləb dxʷ–t’aq’t tiʔiɬ s=ɬalil ʔə tiʔiɬ čxʷəluʔ

immediately CNTRPT–inland DIST NM=go.ashore PR DIST whale

‘right away Whale went way up on shore’

(lit. ‘right away the going-ashore of Whale [was] inland’)

(Hess 1995: 141, line 31)

b. dxʷt’aq’t tiʔəʔ dsuʔibəš

dxʷ–t’aq’t tiʔəʔ d=s=ʔu–ʔibəš

CNTRPT–inland PROX 1SG.PO=NM=PFV–travel

‘where I am traveling [is] inland’

(Hess 1998: 79, line 45)

dxʷ also combines with the question word čad ‘where?’, to form another question word, dxʷčad

‘to where?’:

306

(375) a. dxʷčad tiʔəʔ ʔuqadadid

dxʷ–čad tiʔəʔ ʔu–qada–di–d

CNTRPT–where PROX PFV–stolen–SS–ICS

‘where had the one who stole from him gone to?’

(Hess 2006: 57, line 379)

b. gʷəl ʔux%ʷtubəxʷ ʔə kʷi stab dxʷčadəxʷ

gʷəl ʔux%ʷ–txʷ–b=əxʷ ʔə kʷi stab dxʷ–čad=əxʷ

SCONJ go–ECS–PASS=now PR REM what CNTRPT–where=now

‘and something took him somewhere’

(Hess 2006: 27, line 121)

c. ləʔibəš bək’ʷ dxʷčad

lə=ʔibəš bək’ʷ dxʷ–čad

PROG=travel all CNTRPT–where

‘he traveled to everywhere’

(Hess 1998: 65, line 14)

Like other questions words, dxʷčad can also be used pronominally to mean ‘to somewhere’, as

shown in (375b). It also combines with the adverb bək’ʷ ‘all’ to express the meaning

‘everywhere’ (375c). In addition, dxʷ forms part of the centripetal preposition dxʷʔal ‘to,

toward’, discussed in detail in Section 2.3.1.2 above.

The particle tul’ ‘centrifugal [CNTRFG]’ is part of locative adverbial expressions describing

motion away from a region or location:

(376) a. tux%ʷ čəɬ p’aƛ’aƛ’ tukʷədalikʷ tul’ʔa

tux%ʷ čəɬ p’aƛ’aƛ’ tu=kʷəd–alikʷ tul’–ʔa

just 1PL.SUB worthless PAST=taken–ACT CNTRFG–be.there

‘we just took worthless bits from there’

(Hess 1998: 75, line 261)

b. hədiʔw’əxʷ tiʔəʔ buʔqʷ tul’šqalatxʷ

hədiʔw’=əxʷ tiʔəʔ buʔqʷ tul’–šq•alatxʷ

be.inside=now PROX waterfowl CNTRFG–high• house

‘the waterfowl entered the house from the roof’

(Hess 2006: 75, line 788)

These adverbs are used both as predicate modifiers, as in (376), and as predicates, as in (377):

307

(377) a. tul’t’aq’t tiʔəʔ suʔəƛ’ ʔə tiʔəʔ qʷuʔ

tul’–t’aq’t tiʔəʔ s=ʔu–ʔəƛ’ ʔə tiʔəʔ qʷuʔ

CNTRFG–inland PROX NM=PFV–come PR PROX water

‘it came down to the sea from inland’ (lit. ‘its coming was from inland to the water’)

(Hess 1998: 69, line 108)

b. tul’q’ixʷ tiʔiɬ saʔsaʔ

tul’–q’ixʷ tiʔiɬ saʔ–saʔ

CNTRFG–upstream DIST DSTR–bad

‘it is bad upstream’

[ML Basket Ogress, line 208]

Like its centripetal counterpart dxʷ, tul’ combines with the question word čad ‘where?’ to form

the interrogative tul’čad ‘from where?’:

(378) a. tul’čad kʷi adsuʔibəš sgʷəlub

tul’–čad kʷi ad=s=ʔu–ʔibəš sgʷəlub

CNTRFG–where REM 2SG.PO=NM=PFV–travel pheasant

‘where are you traveling from, Pheasant?’

(Hess 1998: 79, line 43)

b. gʷəl wiliq’ʷitəb tutul’čadəs

gʷəl wiliq’ʷitəb tu=tul’–čad=əs

SCONJ ask–ICS–PASS PAST=CNTRFG–where=3SBJ

‘and they asked him where he might be from’

(Hess 1998: 97, line 166)

As is to be expected, tul’čad has both interrogative and pronominal uses. The centrifugal particle

is also part of the pronoun tul’ʔal ‘from, out of’ (Section 2.3.1.3).

The particle liɬ ‘prolative [PRLV]’ is used to express motion along a path or location

throughout the region designated by its base. Stems formed with liɬ, like their bases, serve most

commonly as predicate modifiers, as in (379):

(379) a. gʷəl x%ʷul’əxʷ čələp p’aƛ’aƛ’ ɬudxʷbəxʷiqad liɬʔilgʷiɬ

gʷəl x%ʷul’=əxʷ čələp p’aƛ’aƛ’ ɬu=dxʷbəxʷiqad liɬ–ʔil•gʷiɬ

then only=now 2PL.SUB worthless IRR=scavenge PRLV–edge•waterway

‘so you guys will just scavenge like good-for-nothings along the shore’

(Hess 1998: 75, line 262)

308

b. huy ʔuʔux%ʷəxʷ liɬt’aq’t

huy ʔu–ʔux%ʷ=əxʷ liɬ–t’aq’t

SCONJ PFV–go=now PRLV–inland

‘he went along the bluff [i.e., just inland of the water]’

(Hess 1998: 68, line 72)

It is also found as part of locative adverbs used as clausal predicates:

(380) a. ck’ʷaqid čəd ɬuliɬšəq

ck’ʷaqid čəd ɬu=liɬ–šəq

always 1SG.SUB IRR=PRLV–high

‘I will always be up top’

[LA Basket Ogress, line 28]

b. yaw’ dᶻəɬ bəliɬč’it kʷi sədxʷluq’ʷucutigʷəds čxʷəluʔ

yaw’ dᶻəɬ bə=liɬ–č’it kʷi s=lə=dxʷluq’ʷucutigʷəd=s čxʷəluʔ

until PTCL ADD=PRLV–be.near REM NM=PROG=sound.in.body=3PO whale

‘as he drew ever closer Whale must have been making sounds in his body’

[ML Mink and Tutyika, I line 24]

and the question word liɬčad ‘which way?’

(381) ɬuliɬčad čəxʷ

ɬu=liɬ–čad čəxʷ

IRR=PRLV–where 2SG.SUB

‘which way will you go?’

(Bates, Hess & Hilbert 1994: 59)

The prolative is also part of the preposition liɬʔal ‘via, through’ (Section 2.3.1.4).

In addition to appearing in regular locative expressions like those in (379), the prolative is

found in some lexicalized forms. Two of these of particular interest are liɬlaq ‘last [in line]’ and

liɬdᶻixʷ ‘first [in line]’ (cf. the forms ʔiɬdᶻixʷ ‘first; better, best’ and ʔiɬlaq ‘later, last’ discussed

in Section 2.1.1.8):

(382) a. tuliɬlaq tsiʔiɬ həbuʔ tiʔiɬ ƛ’usux%aabs

tu=liɬ–laq tsiʔiɬ həbuʔ tiʔiɬ ƛ’u=s=ʔu–x%aab=s

PAST=PRLV–last DIST:FEM pigeon DIST HAB=NM=PFV–cry=3PO

‘Pigeon, who was crying, came last’

(Hess 2006: 32, line 255)

309

b. ʔəƛ’axʷ tiʔiɬ ʔi sbiaw liɬlaq

ʔəƛ’=axʷ tiʔiɬ ʔi sbiaw liɬ–laq

come=now DIST and coyote PRLV–last

‘she and Coyote came last’

(Hess 2006: 35, line 343)

c. diɬ tiʔəʔ skikəwič tiʔəʔ liɬdᶻixʷ ti skʷədatəbs

diɬ tiʔəʔ ski–kəwič tiʔəʔ liɬ–dᶻixʷ ti s=kʷəda–t–əb=s

FOC PROX ATTN–hunchback PROX PRLV–first SPEC NM=taken–ICS–PASS=3PO

‘it was Little Hunchback who was the first one to be taken’

[AJ Basket Ogress, line 42]

As shown by these examples, these two liɬ forms, like the other prolatives illustrated here, are

locative adverbs which treat the relative order in which objects are arranged (i.e., their relative

sequence) in a way analogous to the relative spatial locations expressed by locative adverbs. In

addition, liɬ is used to form three special expressions for location in a canoe — liɬšədᶻt ‘in the

bow’ (from √šədᶻt ‘bow’), liɬʔudəgʷiɬ ‘amidships’ (from ʔudəgʷ- ‘middle’ + -gʷiɬ ‘canoe’, and

liɬʔilaq ‘in the stern’ (from √ʔilaq ‘stern’). In these forms, the prefix seems to have lost its

prolative meaning of ‘via, through’. This also seems to be true of the expression liɬʔalʔal ‘at

home’ (from ʔalʔal ‘house’), as well as the verbs liɬʔa ‘be there’ and liɬʔatxʷ ‘put something

there’ (based on ʔa ‘be there’), although these last two may express nuances that have been lost

in the glosses of the sentences in which they occur.

2.7.5 Interjections and conjunctions

sconj attracting s2 clitics

d. huy čələp ʔuhəliʔdubuɬ ʔə tiʔəʔ shuy čəɬ

huy čələp ʔu–həliʔ–dxʷ–buɬ ʔə tiʔəʔ s=huy čəɬ

SCONJ 2PL.SUB PFV–alive–DC–1PL.OBJ PR PROX NM=be.done 1PL.PO

‘for you are the ones who healed us from what was done to us’

(Hess 2006: 66, line 595)

310

2.8 Compounding

True compounding — the lexicalized combination of one or more radicals into a new lexeme

— is a marginal process is Lushootseed. There seem to be only seven true compound words, four

of them nouns and three of them verbs:

šəɬbadəb ‘stepfather’ (n)

(from √šəɬ ‘make’ + √bad ‘father’ + -əb ‘middle’)

šəɬbədaʔəb ‘stepchild’ (n)

(from √šəɬ ‘make’ + √bədaʔ ‘offspring’ + -əb ‘middle’)

šəɬtadəb ‘stepmother’ (n)

(from √šəɬ ‘make’ + *√tad ‘wife’ + -əb ‘middle’)

šəɬx&əčəb ‘law; will; intention’ (n)

(from √šəɬ ‘make’ + √x&əč ‘mind’ + -əb ‘middle’)

sqəlalitut ‘have a dream’ (v)

(from º√qəl ‘stop’ + -al ‘connective’ + √ʔitut ‘sleep’)

səsaʔalitut ‘have a nightmare’ (v)

(from √saʔ ‘be bad’ + -al ‘connective’ + √ʔitut ‘sleep’)

x&əɬəɬx&əč ‘be sad, be sorry’ (v)

(from √x&əɬ ‘sick’ + Type III reduplication + √x&əč ‘mind’)

Table 74: Compound words

Even these words clearly belong to restricted sets that have obvious precedents in more

productive processes. The first four nouns in this set are based on the verb šəɬ ‘make’,

116

which

in its more frequent use takes a bare nominal predicate complement (see Section 8.2.5 below);

the šəɬ forms in Table 74 seem to be lexicalizations of frequent verb–complement sequences into

a small set of analogous kinship terms. The next two forms, sqəlalitut ‘have a dream’ and

səsaʔalitut ‘have a nightmare’ appear to be compounds of two verbal radicals; however, the

second member of each compound is a reduced from of the radical √ʔitut ‘sleep’, linked to the

previous radical by a connective, -al (to this list might also be added x&ʷalitut ‘snore’, although

the origins of the initial /x&ʷ/ are unknown). This is reminiscent of lexical suffixation (Section

2.1.6), and it may be the case that -itut could be treated as a lexical suffix, one that is

(exceptionally) related etymologically to a synonymous free form. It certainly looks like a lexical

suffix in the making. Only the final form in Table 74, x&əɬəɬx&əč ‘be sad, be sorry’ (also attested as

x&əɬx&əč), seems to be a true compound.

116

This verb is čəɬ in Skagit; all of the š&əɬ- forms in Table 74 would begin with čəɬ- in Skagit.

311

3 Phrase-level clitics

3.1 Tense and mood

mood on a preposition

(383) ƛ’ubəxʷ čəxʷ x%ʷul’əxʷ ɬubəščəb, ɬup’aƛ’aƛ’, ɬusdukʷ ɬuʔal tudiʔ čaʔkʷ

ƛ’ub=əxʷ čəxʷ x%ʷul’=əxʷ ɬu=bəščəb ɬu=p’aƛ’aƛ’ ɬu=sdukʷ

good=now 2SG.SUB only=now IRR=mink IRR=worthless IRR=riff.raff

ɬu=ʔal tudiʔ čaʔkʷ

IRR=at DIST.DMA seaward

‘You will be just a mink, a no-account, riff-raff down there by the water’

(Hess 1998: 69, line 122)

3.1.1 Past tense tu=

b. ƛ’ub čəxʷ ʔubiqʷyitəb ʔə tsiʔacəc tubədaʔs

ƛ’ub čəxʷ ʔu–biqʷyit–b ʔə tsiʔacəc tu=bədaʔ–s

well 2SG.SUB PFV–permit–PASS PR UNQ:FEM PAST=offspring–3PO

‘you should be permitted the deceased’s daughter’

(Hess 1998: 98, line 205)

3.1.2 Habitual mood ƛ’u=

3.1.3 Irrealis mood ɬu=

used in hortatives and polite imperatives

3.1.4 Subjunctive mood gʷə=

softens requests

(384) a. gʷəʔəƛ’txʷ čəxʷ ʔu

gʷə=ʔəƛ’–txʷ čəxʷ ʔu

SBJ=come–ECS 2SG:SUB INT

‘would you bring it?’

(Hess & Hilbert 1976: II, 6)

312

b. gʷət’uk’ʷtubš čələp ʔu dxʷʔal ti xʷuyubalʔtxʷ

gʷə=t’uk’ʷ–txʷ–bš čələp ʔu dxʷ–ʔal ti xʷuyubalʔtxʷ

SBJ=come–ECS–1SG.OBJ 2PL:SUB INT CNTRPT–at SPEC store

‘would you guys take me to the store?’

(Hess & Hilbert 1976: II, 8)

3.1.5 Negative mood lə=

3.2 Progressive clitic lə=

3.3 Additive clitic bə=

Temporal enclitic =axʷ two examples with -\x„ first t(u)asw\li˙ k„i d\x„ßuk’„ilqs\d\x„s ˙\

ti˙i¬ x„(h)udad

tu–˙as–w\li˙ k„i d\x„–ßuk’„–il–qs–\d–\x„s ˙\ ti˙i¬ x„hudad

past–stat–be.visible det np2–grey–inch–nose–ics–now Pr det ash

It showed, so she grayed their noses with ashes [to cover the greasy remains of the

food they had eaten].

seal hunters 51

ß\¬\x„s \lg„\˙ ¬u˙\¬\d

s–ß\¬–\x„–s \lg„\˙ ¬u–˙\¬\d

np–make–now–3po plural irr–eat

They [the brothers-in-law] made them into food for the future.SEAL HUNTERS 492

SHOWS -S AFTER -\X„

BUT (many, many examples of -s first)

d\x„sg„a˙s\x„ g„\¬\x„ sqig„ac ti˙\˙ q’\d¸a≈ ˙\ tsi˙ac\c p’uay’ ˙\ t\ Ò’usuk’„iç’it\bs

d\x„–s–g„a˙–s–\x„ g„\¬–\x„ sqig„ac ti˙\˙ q’\d¸a≈ ˙\ tsi˙ac\c

p’uay’ ˙\ t\ Ò’u–s–˙u–k’„iç’i–t–\b–s

np2–np–be.ones.own–3po–now belong–now deer det intestines Pr

det:fem flounder Pr det hab–np–pfv–butcher–ics–pass–3po

b\sg„a˙s\x„ b\sx„i˙x„i˙s k„\di˙ sucutcuts

b\–sg„a˙–s–\x„ b\–s–x„i˙x„i˙–s k„\di˙ s–˙u–cut–cut–s

add–one’s.own–3po–now add–np–forage–3po det np–pfv–dstr–say–3po

It was his own game he was saying.

That’s why Flounder has Deer’s intestines when she is butchered.

≈„ul’ul’\x„ p’q’ac ˙al k„i tus¬çils\x„

≈„ul’–ul’–\x„ p’q’ac ˙al k„i tu–s–¬çil–s–\x„

exc–only–now rotten.wood Pr det past–np–arrive–3po–now

It was nothing but a rotten log when he arrived.

di¬\x„ ti˙i¬ tud\x„utabs sx„\¬ils\x„

di¬–\x„ ti˙i¬ tu–d\x„–˙u–tab–s s–x„\¬il–s–\x„

idn–now det past–np2–pfv–do–3po np–be.worked.up–3po–now

That is why he did it, [why] he was so worked up.

313

di¬i¬\x„ ti˙i¬ q’\d¸a≈, ¬uq’\d¸a≈s\x„ tsi˙i¬, g„\¬ sqig„ac q’\d¸a≈

di¬–i¬–\x„ ti˙i¬ q’\d¸a≈ ¬u–q’\d¸a≈–s–\x„ tsi˙i¬ g„\¬ sqig„ac q’\d¸a≈

pl–idn–now det intestines irr–intestines–3po–now det:fem ASSC deer

intestines

Those very entrails belonging to Deer will be her entrails

di¬ tushuyut\bs\x„ ˙\ ti˙\˙ duk„ib\¬

di¬ tu–s–huyu–t–\b–s–\x„ ˙\ ti˙\˙ duk„ib\¬

idn past–np–finish–ics–pass–3po–now Pr det Changer

That is what Changer did to her

g„\l (h)uy, ¬çis\x„

g„\l huy ¬çil–s–\x„

conj intj arrive–dir–now

And then he came upon him

g„\l t\¬ biÒ’il ti˙i¬ tud\x„u≈aabs\x„ tsi˙i¬ c\di¬ s˙uß\babdx„

g„\l t\¬ biÒ’–il ti˙i¬ tu–d\x„–˙u–≈aab–s–\x„ tsi˙i¬ c\di¬ s–˙uß\b–a–b–dx„

conj truly smashed–inch det past–np2–pfv–cry–3po–now det:fem he

np–pity–ds1–md–lc

And truly he became squashed which is why the poor dear was crying

314

4 Lexical suffixes

derivation of a bivalent intrans

˙ub\çalq ˙\ t[s]i˙\˙ sqig„ac

˙u–b\ç–alq ˙\ tsi˙\˙ sqig„ac

pfv–fall–game Pr det:fem deer

They brought down deer (Changer line 178)

ʔix&ʷdupt ‘scatter’

ʔix&ʷid ‘throw away’

q’x&ad ‘insult’

q’x&abacəd ‘insult physical appearance’

note that they seem to force schwa epenthesis in CC(V) stems -- k’wcha, tqa, tXwu, etc.

4.1 In verbs

“cognate objects” (also babysitting, addressing as man, etc.)

(385) a. putəxʷ ləq’ʷup’q’ʷup’ačiʔ ʔə tiʔəʔ sčətxʷəd

put–əxʷ lə=q’ʷup’–q’ʷup’•ačiʔ ʔə tiʔəʔ sčətxʷəd

really–now PROG=DSTR–shrivel•hand Pr PROX bear

‘Black Bear’s hands [only] shriveled up [from the heat]’

(Hess 1995: 154, line 81)

b. ləsaxʷəbabac ʔə tiʔəʔ qʷɬaʔy dxʷʔal tiʔiɬ sɬčis tiʔəʔ ʔaʔyəds qaw’qs

117

lə=saxʷəb•abac ʔə tiʔəʔ qʷɬaʔy dxʷ–ʔal tiʔiɬ s=ɬčil–s=s

PROG=jump•body PR PROX log toward–Pr DIST NM=arrive–ALTV=3PO

tiʔəʔ ʔaʔyəd–s qaw’qs

PROX companion–3PO raven

‘he is jumping over the logs until he arrives at his friend, Raven’

(Hilbert & Hess 1977: 25–26)

c. ƛ’utu(u)ʔux%ʷəxʷ ƛ’utušƛ’ap ʔə kʷi sčəbid

ƛ’u=tu=ʔu–ʔux%ʷ=əxʷ ƛ’u=tu=šƛ’•ap ʔə kʷi sčəbid

HAB=PAST=PFV–go=now HAB=PAST=scorch•bottom PR REM fir.bark

‘she would go and she would scorch fir bark’

[DS Star Child, line 173]

117

The word qaw’qs is given in the original source as kaw’qs, although Hess (p.c.) now considers this an error.

315

idiomatic verb with LS

ʔəshuygʷəsəxʷ ʔə tiʔiɬ

ʔəs–huy•gʷəs=əxʷ ʔə tiʔiɬ

stat–be.done•pair=now Pr det

She is married to that one.

4.2 In nouns

4.3 In numeral classification

One function of lexical suffixes is to act as numeral classifiers:

(386) ɬixʷalʔtxʷ ‘three houses’ buusalʔtxʷ ‘four houses’

ɬixʷalgʷiɬ ‘three canoes’ buusalgʷiɬ ‘four canoes’

ɬixʷalps ‘three domestic animals’ buusalps ‘four domestic animals’

ɬixʷalič ‘three bundles’ buusalič ‘four bundles’

ɬixʷəlus ‘three squares (net), stitches’ buusəlus ‘four squares (net), stitches’

ɬixʷilc ‘three dollars’ buusilc ‘four dollars’

118

ɬixʷqs ‘three points’ buusqs ‘four points’

ɬixʷulč ‘three baskets, containers’ buusulč ‘four baskets, containers’

(Hess 1995: 20)

The choice of classifier is largely semantic: lexical suffixes with very concrete, specific

meanings are used in counting those objects which they designate, while others are used when

counting objects that fit into the general class (in either shape or function) of object expressed by

the suffix:

(387) a. ɬixʷagʷiɬ ti dsgʷaʔ

ɬixʷ•agʷiɬ ti d–sgʷaʔ

three•canoe SPEC 1SG.PO–one’s.own

‘I own three canoes’

(Hess & Hilbert 1976: I, 69)

b. ʔəsɬixʷulč ti dsʔax%ʷuʔ

ʔəs–ɬixʷ•ulč ti d–sʔax%ʷuʔ

STAT–three•container SPEC 1SG.PO–clam

‘I have three clams’

(Hess 1995: 20)

118

The lexical suffix -ilc is used in general to refer to round objects; however, when used as a numeral classifier it

can only be used in counting money.

316

The lexical suffix is also used with the interrogative word k’ʷid ‘how many?’:

(388) k’ʷidulč kʷ adsgʷaʔ adyiq’us

k’ʷid•ulč kʷi ad–sgʷaʔ ad–yiq’us

how.many•container REM 2SG.PO–one’s.own 2SG.PO–cedar.root.basket

‘how many cedar-root baskets do you own?’

(Hess & Hilbert 1976: I, 68)

The use of lexical suffixes as numeral classifiers is optional; some lexical suffixes are used in

this way occur only with numbers up to four, while others have an unlimited range.

The use of lexical suffixes as numeral classifiers is only attested in texts when the numeral is

used as a clausal predicate or head of a referential expression rather than as an adnominal

modifier, as in the following examples:

(389) a. x%ʷul’ buusaɬ kʷi sp’ic’ids

x%ʷul’ buus•aɬ kʷi s=p’ic’i–d=s

only four•times REM NM=wrung–ICS=3PO

‘just four times she wrung it out’

[HM Star Child, line 66]

b. tutx%udəxʷ tiʔiɬ ʔə tiʔiɬ buusaɬ

tu=tx%u–d=əxʷ tiʔiɬ ʔə tiʔiɬ buus•aɬ

PAST=stretched–ICS=now DIST PR DIST four•times

‘she pulled four times’

[DS Star Child, line 56]

In elicitation, however, numeral classifier expressions are also used as adnominal modifiers or

the heads of referential expressions:

(390) a. ti ɬixʷgʷiɬ ƛ’əlayʔ

ti ɬixʷgʷiɬ ƛ’əlayʔ

SPEC three•canoe shovel.nosed.canoe

‘three shovel-nosed canoes’

b. ti ɬixʷgʷiɬ

ti ɬixʷgʷiɬ

SPEC three•canoe

‘three canoes’

David Beck 10-2-7 2:19 PM

Comment: classifiers occur on either part of

the numeral

317

In general, expressions of the type shown in (390b) are preferred over the type shown in (a) with

an overt nominal head, (390a) being appropriate only in a situation where it is important to

specify which type of canoe is being counted.

The lexical suffix -aɬ ‘times’ illustrated in (389a) and (b) combines with numerals to form

the base for certain more complex expressions. For instance, adding the inchoative suffix -il

creates verbs indicating the completion of an action a specified number of times:

(391) a. x%ʷul’ ʔubuusaɬil tiʔiɬ shuyuds

x%ʷul’ ʔu–buus•aɬ–il tiʔiɬ s=huyu–d=s

only PFV–four•times–INCH DIST NM=be.done–ICS=3PO

‘she had just done it four times’

[MW Star Child, line 130]

b. ʔal kʷi sɬixʷaɬils, huy, kʷədabacdubutəxʷ tiʔəʔ ʔəskikəwič

ʔal kʷi s=ɬixʷ•aɬ–il=s huy kʷəd•abac–dxʷ–but=əxʷ

PR REM NM=three•times–INCH=3PO SCONJ take•body–DC–REFL=now

tiʔəʔ ʔəs–ki–kəwič

DIST STAT–ATTN–hunchbacked

‘on the third time, well, the little hunchbacked one caught his body [on the snags]’

[AJ Basket Ogress, line 57]

Similar constructions are formed from numeral bases combined both with -aɬ and -dat ‘day’:

(392) ʔaləxʷ sbuusaɬdatils kʷi suq’ilaʔkʷčups sp’ip’ic’ikʷ

ʔal=əxʷ s=buus•aɬ•dat–il=s kʷi s=ʔu–q’il•aʔkʷ•čup=s

PR=now NM=four•times•day–INCH=3PO REM NM=PFV–aboard•group•fire=3PO

sp’ip’ic’ikʷ

Diaper.Child

‘[it was] on the fourth day of Diaper Child’s loading firewood’

[HM Star Child, line 102]

These expressions indicate the time in days that an event lasted or the number of days on which a

repeated event occurred.

NOTE: WILL PROBABLY RE-ANALYZE -ALH ‘CLASSIFIER’ AS A LEXICAL

SUFFIX, ≠ ‘TIMES’

318

Two of the interrogatives, stab ‘what?” and k’ʷid ‘how many’, are regularly found associated

with lexical suffixes acting as classifiers indicating the type of item being asked about. Of the

two, stab seems to combine less freely with the lexical suffixes, having only three forms —

stabaʔkʷəbixʷ ‘what group of people?’, stabac ‘what kind of tree?’, stabidup ‘what kind?’ —

in addition to the generic construction with the classifier, stabaɬ ‘what kind?’. k’ʷid, on the other

hand, combines with a much wider (perhaps unrestricted) range of the lexical suffixes, giving

forms such as k’ʷidalps ‘how many animals?’, k’ʷidaʔltxʷ ‘how many houses?’, k’ʷidalq ‘how

much game?’, k’ʷidgʷiɬ ‘how many canoes?’, and k’ʷidilc ‘how much money?’.

319

5 Reduplication

Reduplication in Lushootseed is a highly productive derivational process applicable to stems

of a variety of lexical classes. Like many Salishan languages, Lushootseed has more than one

reduplicative pattern, and different reduplicative patterns express different meanings — in total,

there are three regular, productive reduplications, as well as two more restricted patterns, and

these are used to express six different meanings. Two of the regular patterns, Type I C

reduplication (Section 5.1) and Type II C

V- reduplication (5.2), each express a single meaning

— ‘distributive’ and ‘attenuative’, respectively. On the other hand, Type III -VC

reduplication

(5.3) is associated with four separate meanings — ‘diminished effectiveness’, ‘intensivity’,

‘exclusivity’, and ‘plurality’, depending on the lexical and semantic class of the base it is applied

to. There are also two minor patterns, C

$- and -V

- reduplication, used to express the plural of

lexicalized sets of words. These will be discussed in Section 5.4, followed by an examination of

the combinatorial potential of the different reduplicative patterns, some of which can co-occur on

a single base (5.5). As will become clear in the discussion that follows, reduplicative processes in

Lushootseed are very complex, and the current situation in the synchronic grammar may reflect

an earlier stage of the language in which even more reduplicative patterns existed, some of which

have become neutralized and collapsed into single semantic categories, with some individual

words retaining idiosyncrasies inherited from what was once probably a distinct reduplicative

process. Consequently, even for the most regular reduplications, there are many irregular forms,

and there are many variations on the basic patterns that apply to what appear to be lexically-

specified sets of stems. There have been a number of attempts in the literature to come to grips

with these apparent irregularities (e.g., Bates 1986; Urbanczyk 1996a), and these have managed

quite successfully to account for a number of the apparent irregularities in the system; however, a

lot of this work rests on complex formal analyses that are beyond the scope of a descriptive

grammar, and even the best approach still requires the specification of a large number of

David Beck 10-2-7 2:19 PM

Comment: don't forget the reduplicative

patterns of /ah and qah and other "h" final

radicals

320

lexically-specified exceptions to the underlying generalizations. Rather than being presented in

abstract terms or in the context of a theoretical model, the data in the discussion that follows will

simply be presented in sets of related forms; the reader interested in pursuing the patterns further

and untangling the on-going debates on reduplicated forms in Lushootseed is referred to the

theoretical literature.

5.1 Type I reduplication C

- ‘distributive’

One of the most common and regular of the reduplicative patterns is the Type I distributive

[DSTR] reduplication. The semantic effect of Type I reduplication varies slightly depending on

the lexical class of the stem. With nouns, this reduplicative pattern is the most common way of

indicating plurality; with verbs, it indicates either locative distributivity or iteration of an action.

In a few cases, it seems primarily to function as an indication of subject plurality, perhaps relying

on the implication that an action performed by more than one person is done more than once in

more than one location. The specific reading given to a verb that has undergone Type I

reduplication is to a certain extent context-dependent, though all possible readings fit neatly

under the abstract notions of locative and/or temporal distributivity.

Type I reduplication copies the first three segments of the base as a prefix:

(393) b$daʔ ‘offspring, child’ > b$db$daʔ ‘children’

cut ‘speak’ > cutcut ‘say over and over’

dᶻ$kʷ ‘wander’ > dᶻ$kʷdᶻ$kʷ ‘wander all over’

haʔɬ ‘good’ > haʔhaʔɬ ‘many are good’

gʷədil ‘sit down’ > gʷədgʷədil ‘sit down all over (group)’

qʷatad ‘lay something down’ > qʷatqʷatad ‘lay something down all over’

suq’ʷaʔ ‘younger brother’ > suq’ʷsuq’ʷaʔ ‘younger brothers’

tux%ʷud ‘stretch something’ > tux%ʷtux%ʷud ‘stretch something out’

tuʔad ‘spit something out’ > tuʔtuʔad ‘spit something out over and over’

x%aƛ’is ‘fight something off’ > x%aƛ’x%aƛ’is ‘fight one thing off then another’

yubil ‘starve (human)’ > yubyubil ‘be starving all over (human)’

119

This verb is also used to mean ‘die’ when referring to animals.

321

If the first syllable of the base contains a long vowel, the second mora of the long vowel is

copied instead of the coda consonant, as in x&aabiʔɬ ‘cry (baby)’ > x&aax&aabiʔɬ ‘cry and cry

(baby)’. If the stem being reduplicated begins with a prefix, the prefix is ignored and

reduplication starts at the left edge of the root (e.g., dxʷləgʷləgʷəb ‘youths’, ʔəbsqʷəbqʷəbayʔ

‘have dogs’). Note that this includes synchronically unanalyzable prefixes such as the initial /s/

in words like sqʷəbayʔ ‘dog’ (cf. sqʷəbqʷəbayʔ ‘dogs’).

When the first syllable of the base ends in /l/, the final segment of the reduplicand is

glottalized, as in (394):

(394) q$ladiʔ ‘snag, uprooted tree’ > q$l’q$ladiʔ ‘snags’

saliʔ ‘two’ > sal’saliʔ ‘two by two’

sgʷəlub ‘pheasant’ > sgʷəl’sgʷəlub ‘pheasants’

stuləkʷ ‘river’ > stul’tuləkʷ ‘rivers’

təlawil ‘run’ > təl’təlawil ‘run and run’

yəlab ‘sibling of deceased person’ > yəl’yəlab ‘ancestors, parents’

This may apply to other approximants as well, but attested examples are not found in the present

corpus (see, however, the glottalization of approximants in some Type II reduplications,

discussed in Section 5.2 below).

With C$C radicals that have been transitivized with the [-əd] allomorph of the internal

causative suffix -t (Section 2.1.2.1), Type I reduplication causes the schwa of the suffix to be

realized as /a/:

(395) b$q’$d ‘swallow something’ > bəq’bəq’ad ‘many swallow something’

dəšəd ‘set something on its side’ > dəšdəšad ‘many set somethings on their sides’

tədəd ‘put somethings in a row’ > tədtədad ‘many put somethings in rows’

xʷəbəd ‘throw something away’ > xʷəbxʷəbad ‘many throw somethings away’

The same applies to the internal causative forms of (non-epenthesizing) CC radicals (see Section

2), which also add an epenthetic schwa to the reduplicand:

(396) šəqəd ‘raise something’ > šəqšqad ‘raise somethings’

t’əq’əd ‘patch something’ > t’əq’t’q’ad ‘many patch somethings all about’

x%əƛ’ ‘bite something’ > x%əƛ’x%ƛ’ad ‘many bite something’

322

The schwa-epenthesis in the reduplicand is probably attributable both to the need to break up a

lengthy consonant cluster and the need to conserve the canonical CVC shape of the reduplicand.

5.2 Type II reduplication C

- ‘attenuative’

Another common reduplicative pattern is the Type II attenuative [ATTN] reduplication. With

nominal bases, the meaning of this type of reduplication is that of a diminutive, expressing

smallness in size and/or endearment. The meaning of this reduplication with verbal bases is a bit

more diffuse, though it generally conveys the notion of a diminished action, one that is only just

or partially completed, effected with less than normal force, or whose results are of reduced size

or scope. With verbs in particular, the readings of Type II reduplications are highly-context

specific and, in many of the textual attestations, the import of the reduplication has been lost or

blurred by translation. This is of particular concern because many of the bases for reduplication

discussed below have more than one form, and without more precise glossing and the intuitions

of native speakers it is not always possible to be sure that these are synonymous forms rather

than different lexical items created by slightly different reduplicative patterns.

In comparison with Type I reduplication, Type II is more variable and seems to be a part of a

greater number of idiosyncratic, lexicalized stems. In its simplest form, Type II reduplication

copies the first two segments of the base as a prefix:

(397) ʔəsɬič’txʷ ‘have something cut up’ > ʔəsɬiɬič’txʷ ‘have something cut up into bits’

ʔəsƛ’ubil ‘become well’ > ʔəsƛ’uƛ’ubil ‘become a bit better’

ʔistaʔb ‘be like’ > ʔiʔistaʔb ‘be a bit like’

ʔux%ʷ ‘go’ > ʔuʔux%ʷ ‘go on and on’

badil ‘mountain’ > babadil ‘little mountain’

č’it ‘be near’ > č’ič’it ‘be just here’

diʔucid ‘opposite side of river’ > didiʔucid ‘just across the river’

luƛ’ ‘be old’ > luluƛ’ ‘be a bit old’

ƛ’ac’apəd ‘belt’ > ƛ’aƛ’ac’apəd ‘small belt; ant’

ƛ’iq’šəd ‘sticky-foot’ > ƛ’iƛ’iq’šəd ‘sapsucker (lit. ‘little sticky-foot’)’

ƛ’uil ‘grow thin’ > ƛ’uƛ’uil ‘grow a bit thin’

laq ‘be behind’ > lalaq ‘be a bit behind’

ləliʔxʷ ‘manage to change it’ > lələliʔxʷ ‘just manage to change it’

323

luƛ’əb ‘grow old’ > luluƛ’əb ‘grow a bit older’

saq’ʷ ‘fly’ > sasaq’ʷ ‘take flight’

saxʷəb ‘jump, sprint’ > sasaxʷəb ‘scamper off’

stubš ‘man’ > stutubš ‘boy’

suqwaʔ ‘younger sibling or cousin’ > susuqwaʔ ‘little younger sibling or cousin’

t’agʷt ‘be on top’ > t’at’agʷt ‘be just on top’

wiliq’ʷid ‘ask someone’ > wiwiliq’ʷid ‘quiz someone’

xilix%txʷ ‘wage war on’ > x%ix%ilix%txʷ ‘compete with’

As with Type I reduplication, the reduplicand is added directly to the root, ignoring any prefixes,

such as the stative prefix ʔəs- in the first two forms in (397). This also includes synchronically

unanalyzable instances of the nominalizing prefix, s- — e.g., stubš ‘man’ > stutubš ‘boy’.

Frequently, Type II reduplication is accompanied by vowel reduction either in the base or in

the reduplicand. In the following forms, the first vowel of the root is reduced to schwa:

(398) ʔayil ‘hide’ > ʔaʔəy’il ‘skulk around’

čalal’ ‘be short of goal’ > čičəlal’ ‘be just short of goal’

k’ʷid ‘amount’ > k’ʷik’ʷəd ‘small amount’

q’ilaʔkʷčup ‘load wood on canoe’ > q’iq’əlaʔkʷčup ‘load wood on canoe’

studəq ‘slave’ > stutədəq ‘little slave’

stuləkʷ ‘river’ > stutələkʷ ‘creek’

t’agʷtaptxʷ ‘place end on top’ > t’at’əgʷtaptxʷ ‘support part of load on top’

waq’waq’ ‘frog’ > waw’əq’wəq’ ‘little frog’ (Sk)

xilix%txʷ ‘wage war on’ > x%ix%əlix%txʷ ‘pick a fight with’

yubil ‘animal dies’ > yuyəbil ‘small animal dies’

Which bases undergo this reduction is not entirely predictable, although it has been linked to

prosodic conditions and sonority (Bates 1986; Urbanczyk 1996a). The form ʔaʔəy’il ‘skulk

around’ glottalizes the /y/ of its base, ʔayil ‘hide’, as well as reducing the vowel. Note also the

form waw’əq’wəq’ ‘little frog’ (from waq’waq’ ‘frog’), which glottalizes the initial /w/ of the

base and reduces both vowels to schwa (see also the form waw’lis in 399 below). This form is

from a Skagit speaker — the attested form in Snohomish texts is waw’q’waq’. One form in this

list, x&ix&əlix&txʷ ‘pick a fight with’, also appears with a slightly different gloss as x&ix&ilix&txʷ

‘compete with’ in the list in (397) of forms that do not undergo reduction.

Another set of reduplicated forms removes the first vowel of the base entirely:

(399) ʔ$sqʷatx, ‘have laid out’ > ʔ$sqʷaqʷtx, ‘have laid out’

324

č’it ‘be near’ > č’ič’t ‘be just here’

k’ʷat’ad ‘mouse’ > k’ʷak’ʷt’ad ‘little mouse’

ƛ’iq ‘emerge’ > ƛ’iƛ’q ‘emerge intermittently’

pastəd ‘white person’ > papstəd ‘white child’

sɬadəyʔ ‘woman’ > sɬaɬdəyʔ ‘girl’

šukʷild ‘grey something’ > šuškʷild ‘grey something a bit’

t’isəd ‘arrow’ > t’it’səd ‘little arrow’

t’uc’il ‘shoot something’ > t’ut’c’il ‘shoot a little something’

walis ‘frog’ > waw’lis ‘little frog’

waq’waq’ ‘frog’ > waw’q’waq’ ‘little frog’ (Sn.)

As noted in Hess & Hilbert (1976) and Hess (1998), the majority of forms that undergo this kind

of syncope have bases with a voiceless consonant in both the onset and the coda of the first

syllable of the base. However, there are some exceptions to this generalization (e.g., waw’q’waq’

‘little frog’), and not all bases of this form undergo syncope in Type II reduplication (cf.,

ƛ’aƛ’ac’apəd ‘small belt; ant’ and stutələkʷ ‘creek’; also, note the form č’ič’it ‘be just here’,

which co-exists with č’ič’t ‘be just here’). Both Bates (1986) and Urbanczyk (1996a) point to

sonority, stress-assignment, and other prosodic factors as conditions on syncope in this

reduplicative pattern.

In at least two cases, the same base gives rise to separate lexemes, depending on whether the

Type II reduplication is applied with or without syncope:

(400) papst$d ‘white child’ vs. papastəd ‘white person [derogatory]’

sɬaɬdəyʔ ‘girl’ vs. sɬaɬədəyʔ ‘girlfriend’

120

The first pair of examples shows a contrast between Type II reduplication triggering syncope of

the first vowel of the base and Type II reduplication that leaves the base intact. The second pair

contrasts a form showing syncope of the base with another form that reduces the base vowel to

schwa.

There are also a few forms where the attenuative reduplicand is C

ə-:

(401) gʷad ‘speak’ > gʷ$gʷad ‘chit-chat’

gʷaʔtxʷ ‘accompany someone’ > gʷəgʷaʔtxʷ ‘join someone’

q’x&abacəd ‘insult someone’s looks’ > q’əq’x&abacəd ‘comment on someone’s looks’

120

Cf. ʔiɬsɬəɬadəyʔ ‘she who of the others is your woman’, formed with the partitive prefix.

325

staləɬ ‘nephew, niece’ > stətaləɬ ‘young nephew or niece’

šiqʷ ‘hat’ > šəšiqʷ ‘favourite hat’

While this list is not extensive, some of these lexical items are of relatively high frequency in the

spoken language.

Another common variation on Type II reduplication uses C

i- as the reduplicand rather than

-. This pattern is predictable for most bases with initial CC clusters:

(402) cqʷuɬ ‘post’ > ʔəscicqʷuɬ ‘be sort of propped up’

č’ƛ’aʔ ‘stone’ > č’ič’ƛ’aʔ ‘little stone’

gʷəƛ’əlad ‘stop’ > gʷəƛ’iƛ’əlad ‘pause’

121

ɬčil ‘arrive’ > ɬiɬčil ‘arrive near goal’

q’x,ad ‘insult someone’ > q’iq’x,ad ‘call someone names’

qʷɬayulč ‘wooden dish’ > qʷiqʷɬayulč ‘small wooden dish’

qʷɬayʔ ‘log’ > qʷiqʷɬayʔ ‘stick’

tx&ud ‘stretch something’ > titx&ud ‘stretch something a bit’

x&q ‘be wrapped’ > x&ix&q ‘wrapped up a little’

There are, however, a number of other words which take a C

i- Type II reduplicand:

(403) b$č ‘be fallen’ > bib$č ‘drop by’

buʔqʷ ‘waterfowl’ > bibuʔqʷ ‘small, useless bird’

c$lac ‘five’ > cic$lac ‘five small items’

č$gʷas ‘wife’ > čič$gʷas ‘dear wife’

daʔac ‘call me’ > didaʔac ‘call lil’ ol’ me’

gʷədil ‘sit down’ > gʷigʷədil ‘sit down briefly’

haʔɬ ‘good’ > hihaʔɬ ‘nice little’

k’ʷəɬ ‘be poured out’ > k’ʷik’ʷəɬ ‘be trickling a bit’

puʔtəd ‘shirt’ > pipuʔtəd ‘favourite shirt, thin shirt, baby’s shirt’

p’aʔcut ‘try to do it’ > p’ip’aʔcut ‘try to do it over and over’

p’uay’ ‘flounder’ > p’ip’uay’ ‘little flounder’

qaw’qs ‘raven’ > qiqaw’qs ‘fledgling raven’

sgʷəlub ‘pheasant’ > sgʷigʷəlub ‘fledgling pheasant’

skəwič ‘hunchback’ > kikəwič ‘Little Hunchback (name)’

sq’ʷəlaɬəd ‘berry’ > sq’ʷiq’ʷəlaɬəd ‘little berry’

təbɬusəd ‘put ochre on face’ > titəbɬusəd ‘put a bit of ochre on face’

təlawil ‘run’ > titəlawil ‘jog’

The only predictable members of this second list are those bases which have a schwa as the first

vowel; however, there are a few stems which have a schwa in the first syllable that either take a

- reduplicand (e.g., lələliʔ ‘just manage to change it’) or a C

iʔ- reduplicand (biʔbədaʔ

121

The first syllable in these forms is the prefix gʷə- ‘dubitative’ (Section 2.1.6).

326

‘small offspring’) (see below). A few stems combine C

i- reduplication with syncope of the first

vowel of the base:

(404) ʔ$ɬ$d ‘feed on something’ > ʔiʔɬ$d ‘eat a bit of something’

b$šč$b ‘mink’ > bibšč$b ‘little mink’

d$č’uʔ ‘one’ > didč’uʔ ‘one small item; solitary’

Another common variant on C

- reduplication is C

ʔ- reduplication:

(405) ɬaɬlil ‘live there’ > ɬaʔɬaɬlil ‘a few live there’

lil ‘be near’ > liʔlil ‘be a little ways off’

luƛ’ ‘be old’ > luʔluƛ’ ‘be a bit old’

mad ‘be small’ > maʔmad ‘be a bit small’

saxʷ$b ‘jump, sprint’ > saʔsaxʷəb ‘scamper off’

suqʷaʔ ‘younger sibling or cousin’ > suʔsuqʷaʔ ‘little younger sibling or cousin’

talə ‘money’ > taʔtalə ‘a bit of money’

x%ayəb ‘laugh’ > x%aʔx%ayəb ‘giggle’

Which bases undergo this type of reduplication is inherently unpredictable, and at least two

words in this list — saxʷəb ‘run, sprint’ and luƛ’ ‘be old’ — also undergo plain C

reduplication without any obvious difference in meaning.

122

Like C

- reduplicands, C

ʔ-

reduplicands can also trigger reduction or syncope of the first vowel in the base:

(406) ɬid ‘be tied’ > ɬiʔɬdah$b ‘trawl’

k’ʷilid ‘peek at something’ > k’ʷiʔk’ʷ$l ‘take a little peek at something’

saliʔ ‘two’ > saʔs$liʔ ‘two little items’

saq’ʷ ‘fly’ > saʔsq’ʷ ‘fly just a bit’

saxʷəb ‘jump, sprint’ > saʔsxʷəb ‘run a few steps’

six%ʷid ‘make noise moving water’ > siʔsx%ʷicut ‘go along in water making noise’

stubš ‘man’ > stuʔtəbš ‘one man among women’

stuləkʷ ‘river’ > stuʔtələkʷ ‘creek’

Of these forms, saʔsq’ʷ ‘fly just a bit’, stuʔtuləkʷ ‘creek’, stuʔtəbš ‘one man among women’,

and saʔsxʷəb ‘run a few steps’ have attested forms undergoing plain C

- reduplication without

the glottal stop — sasaq’ʷ ‘take flight’, stutələkʷ ‘creek’, stutubš ‘boy’, and sasaxʷəb ‘scamper

off’, respectively. Note that there are various reduplicated forms of saxʷəb ‘jump, sprint’ —

sasaxʷəb/saʔsaxʷəb ‘scamper off’, saʔsxʷəb ‘run a few steps’, and səʔsxʷab ‘hop’. The same is

122

saxʷəb is also attested with the reduplicated form səʔsxʷab, glossed as ‘hop, Indian broad jump (game)’, which

may be based diachronically on a now-unattested base *sxʷab.

327

true of stubš ‘man’ which has two different reduplicated forms, stutubš ‘boy’ and stuʔtəbš ‘one

man among women’, which clearly have different meanings and belong to different lexemes. In

the case of sasaq’ʷ ‘take flight’, which is given a slightly different gloss than saʔsq’ʷ ‘fly just a

bit’, however, it is not clear in this case to what degree the difference in gloss is due to context or

to the fact that the two words belong to different lexemes.

ʔ- reduplication also has a C

iʔ- variant:

(407) b$daʔ ‘offspring’ > biʔb$daʔ ‘little child’

b$lx,ʷəd ‘pass someone’ > biʔbəlx%ʷəd ‘pass someone by a little bit’

buus ‘four’ > biʔbuus ‘four small items’

gʷəq’əd ‘open something’ > gʷiʔgʷəq’əd ‘open something briefly’

luud ‘hear something’ > liʔluud ‘hear a little bit of something’

ɬx%ʷubx% ‘sister of Basket Ogress’ > ɬiʔɬx%ʷubx% ‘sister of Basket Ogress’ (diminutive)

p’uay ‘flounder (fish)’ > p’iʔp’uay ‘little flounder’

qəladiʔ ‘uprooted tree, snag’ > qiʔqəladiʔ ‘Little Uprooted Tree (name)’

qʷuʔ ‘fresh water’ > qʷiqʷuʔ ‘small amount of freshwater’

sdəxʷiɬ ‘hunting canoe’ > sdiʔdəxʷiɬ ‘little hunting canoe’

sqʷəbayʔ ‘dog’ > sqʷiʔsqʷəbayʔ ‘puppy’

sx%ʷədiʔ ‘bullhead (fish)’ > sx%ʷiʔx%ʷədiʔ ‘little bullhead’

tsadiʔ ‘hit something on the side’ > tiʔtsadiʔ ‘tap something on the side’

As with plain C

i- reduplication, C

iʔ- reduplication is largely predictable for bases derived from

CC radicals (e.g., tiʔtsadiʔ ‘tap something on the side’ from √ts ‘hit something with fist’), as

well as those beginning with consonant clusters (ɬiʔɬx&ʷubx& ‘little sister of Basket Ogress’). One

of these forms, p’iʔp’uay’ ‘little flounder’, also has a plain C

i- form, p’ip’uay’.

The existence of so many apparently synonymous Type II reduplicated forms, as well as the

unpredictability of which forms consistently require a glottal stop and which do not, has

prompted some interesting theoretical work. In particular, Bates (1986) suggests the

presence/absence of the glottal stop is due to prosodic factors and offers some arguments in

support of this based on words that have synonymous C

- and C

ʔ- reduplicated forms.

However, many words that undergo C

ʔ- reduplication are very frequent and are never attested

without a glottal stop in spite of appearing in a wide variety of prosodic environments. It may

328

well be the case that for some words C

ʔ- (or C

iʔ-) reduplication is partially or wholly

prosodically-conditioned and for others it is lexically-specified. This seems even more likely in

that there are a few cases like stutubš ‘boy’ vs. stuʔtəbš ‘one man among women’ in which

- and C

ʔ- reduplications give clearly different meanings, perhaps an indication that these

once were distinct reduplicative patterns in the language. These may have become neutralized for

most words, giving rise to competing forms, and opening the door to a certain amount of

variation across dialects and speakers. Likewise, vowel reduction and syncope in the base may

be predictable in many cases, as suggested both by Bates (1986) and Urbanczyk (1996a), but it

seems likely that at least some of these variations on Type II reduplication are, in fact, lexically-

specified for their stems. Examples such as those in (400) may indicate that these, too, once

represented separate patterns that collapsed during the historical development of the language.

5.3 Type III reduplication -V

Type III -V

reduplication involves infixing the first vowel and second consonant of the

base after the first consonant. In comparison with other reduplications, especially Type II, Type

III reduplication is formally very regular and presents few phonological complications. On the

other hand, unlike Types I and II, Type III reduplication is not characterizable as expressing a

single meaning or meanings belonging to a coherent semantic field: instead, it is associated with

a variety of meanings, depending in part on the lexical and semantic class of the word it is

applied to.

5.3.1 Diminished effectiveness

When applied to verbal bases expressing actions, Type III reduplication creates expressions

of diminished effectiveness [DIM.EFF] — that is, expressions of activities performed randomly,

ineffectively, or inconclusively, or of languid states:

(408) ʔib$š ‘walk, journey’ > ʔibib$š ‘meander’

329

ʔuluɬ ‘travel by water’ > ʔululuɬ ‘paddle around’

č’axʷacut ‘hit onself, club oneself’ > č’axʷaxʷacut ‘thrash about’

dᶻaq’ ‘topple over’ > dᶻaq’aq’ ‘teeter’

gʷədil ‘sit down’ > gʷədədil ‘just sit around’

k’ʷalčcut ‘bend oneself backwards’ > k’ʷalalčcut ‘bend onself back a bit’

k’ʷil ‘peer out from behind’ > k’ʷilil ‘peer about’

ƛ’iq ‘emerge’ > ƛ’iqiq ‘emerge intermittently’

labəd ‘see something’ > lababəd ‘look around for something’

ɬild ‘give food’ > ɬilild ‘give a bit of food’

saqʷ ‘fly’ > saqʷaqʷ ‘wheeling in the sky’

saxʷəb ‘jump, sprint’ > saxʷaxʷəb ‘scurry about ineffectively’

tab ‘do’ > tabab ‘carry on’

tašad ‘fix up a bit’ > tašašad ‘fix something up a bit’

wəliʔil ‘become visible’ > wələliʔil ‘become visible in glimpses’

x%əɬ ‘be sick’ > x%əɬəɬ ‘be depressed’

123

yubil ‘human starves’ > yububil ‘body is run down, feel unwell’

Judging by the glosses of a few of these forms, ‘diminished effectiveness’ may overlap

somewhat with the attenuative (e.g., ƛ’iqiq ‘emerge intermittently’ vs. ƛ’iƛ’q ‘emerge

intermittently’; see also forms in both categories glossed as ‘X a bit’), though this may have

more to do with the English translations than with the actual meaning of the forms. Most forms

that have both an attested Type II and a Type III reduplication show the predicted difference in

meaning (saxʷaxʷəb ‘scurry about ineffectively’ vs. sasaxʷəb ‘scamper off’; k’ʷilil ‘peer about’

vs. k’ʷiʔk’ʷəl ‘take a little peek at something’; saqʷaqʷ ‘wheeling in the sky’ vs. sasaq’ʷ ‘take

flight’).

5.3.2 Intensivity

When applied to certain adverbs and certain verbs expressing states and locations, Type III

reduplication expresses intensivity [INTNS]:

(409) ʔa ‘be there’ > ʔaʔa ‘be right here’

b$k’ʷ ‘all’ > b$k’ʷ$k’ʷ ‘every last one’

dah ‘be thankful’ > dahah ‘be very thankful’

day’ ‘especially, very, only’ > day’ay’ ‘all, completely, only’

diɬ ‘just that one’ > diɬiɬ ‘just that very one’

qah ‘a lot’ > qahah ‘really a lot’

123

This reduplicated form is also found in the compound x&əɬəɬx&əč ‘feel sad, sorry’ (lit. ‘a-bit sick mind (x&əč)’).

David Beck 10-4-14 1:27 PM

Comment: remove /h/, explain epenthesis

330

tiləb ‘immediately’ > tililəb ‘suddenly, instantly’

tux%ʷ ‘merely’ > tux%ʷux%ʷ ‘not much at all’

x%ʷul’ ‘only, just’ > x%ʷul’ul’ ‘nothing but, entirely’

In addition to these forms, there is diʔɬiɬ ‘thereupon, just as soon as’, formed from a slightly

irregular Type III reduplication of √diʔɬ ‘all at once, suddenly’ which ignores the glottal stop in

the non-reduplicated form, as well as hagʷagʷəxʷ ‘finally, at last’, formed from √haʔkʷ ‘for a

long time’. This radical regularly voices its final consonant and loses the glottal stop when in

non-final position (Section 2). The adverb dᶻixʷ ‘first, before’ has a regularly formed but

idiomatic Type III reduplication — dᶻixʷixʷ ‘even more so’ — which fits semantically with the

category of intensivity but doesn’t have a transparent semantic relationship to the meaning of its

base. There is also a form hayayəd ‘figure out how to deal with something’ based on the bound

radical º√hay ‘be known’ via an unattested intermediate form *hayəd ‘know something’.

124

Finally, the radical √təɬ ‘be true’ has a form təɬəɬtəɬ ‘be really true’ that seems to combine Type

I and Type III reduplication.

5.3.3 Exclusivity

When applied to a handful of pronouns and at least one noun, Type III reduplication has the

effect of expressing exclusivity [EXC] or uniqueness:

(410) c$diɬ ‘he, she’ > cədədiɬ ‘just him, just her’

dibəɬ ‘we’ > dibibəɬ ‘just us’

gʷəlapu ‘you guys’ > gʷələlapu ‘just you guys’

sɬadəyʔ ‘woman’ > sɬadadəyʔ ‘woman living alone’

This meaning is clearly related to the intensifying meaning of Type III reduplications seen with

adverbs and locative verbs.

124

There is, however, a form hay’əd ‘pay attention to’ which may be the synchronic reflex of *hayəd.

331

5.3.4 Plurality

The fourth use of Type III reduplication is to create the plurals of two determiners and a few

lexically-specified nouns:

(411) ʔalš ‘cross-sex sibling’ > ʔalalš ‘cross-sex siblings’

ʔibac ‘grandchild’ > ʔibibac ‘grandchildren’

siʔab ‘noble person’ > siʔiʔab ‘noble people’

125

stubš ‘man’ > stububš ‘men’

tiʔ$ʔ ‘this’ > tiʔiʔ$ʔ ‘these’

tiʔiɬ ‘that’ > tiʔiʔiɬ ‘those’

Type III reduplication is also used as a classificatory morpheme with numerals other than saliʔ

‘two’ and buus ‘four’ when counting humans (see also Table 71 in Section 2.7.2 above):

(412) ɬixʷ ‘three’ > ɬixʷixʷ ‘three people’

c$lac ‘five’ > cələlac ‘five people’

yəlaʔc ‘six’ (NL) > yələlaʔc ‘six people’

dᶻəlačiʔ ‘six’ (SL) > dᶻəlalačiʔ ‘six people’

c’ukʷs ‘seven’ > c’ukʷukʷs ‘seven people’

tqačiʔ ‘eight’ > təqqačiʔ ‘eight people’

x%ʷəl ‘nine’ > x%ʷələl ‘nine people’

ʔulub ‘ten’ (NL) > ʔululub ‘ten people’

padac ‘ten’ (SL) > padadac ‘ten people’

salačiʔ ‘twenty’ > salalačiʔ ‘twenty’ (SL)

Note that the form təqqačiʔ ‘eight people’ treats the initial consonant cluster of tqačiʔ ‘eight’ as

the first syllable of the base and uses a schwa as the vowel for the -V

reduplicand. Type III

reduplication expressing plurality is also applied to the interrogative word k’ʷid ‘how many?’

(Section 8.4.2) to form k’ʷidid ‘how many people?’.

5.4 Other plural reduplications

In addition to using Type I and Type III reduplications to express plurality, Lushootseed

expresses the plurality of a number of words with idiosyncratic reduplicative patterns. One such

group of words takes a prefixal C

$- reduplicand:

(413) dᶻ$dis ‘tooth’ > dᶻ$dᶻ$dis ‘teeth’

126

125

The form siʔiʔab ‘noble people’ is also attested with the meaning ‘head man’ (i.e., ‘most noble’), representing an

intensive interpretation of the same reduplication (ML SH 576).

332

sɬad$yʔ ‘woman’ > sɬ$ɬad$yʔ ‘women’

stud$q ‘slave’ > st$tud$q ‘slaves’

syaʔyaʔ ‘relative, friend’ > sy$yaʔyaʔ ‘relatives, friends’

tagʷ$xʷ ‘hunger’ > t$tagʷ$xʷ ‘many are hungry’

yubil ‘human starves’ > yəyubil ‘many humans starve’

Note that this list contains both verbs and nouns: with nouns, this reduplication expresses plural

number; with verbs, it expressed the plurality of the subject. To the list in (413) we might also

add qʷqʷatil ‘many are laid out’ (from √qʷat ‘be lying; snow falls’), which seems to apply

syncope to the schwa in the reduplicand, and the numerals missing from the list of numeral

forms in (412) used for counting humans (cf. Table 70, Section 2.7.2) — səsaʔliʔ ‘two people’

and bəbuʔs ‘four people’, which both epenthesize a glottal stop after the nucleus of the first

syllable in their base.

Another plural reduplicative pattern involves the doubling of the first vowel in the base:

(414) c$diɬ ‘he, she’ > caadiɬ ‘they’

č$gʷas ‘wife’ > čaagʷ$s ‘wives’

č’ač’as ‘child’ > č’aač’as ‘children’

č’$bas ‘in-law’ > č’aab$s ‘in-laws’

127

gʷ$dil ‘sit’ > gʷaadil ‘two or more sit’

siʔab ‘noble person’ > siiaʔb ‘noble people’

š$dᶻal ‘go outdoors’ > šaadᶻal ‘many go outdoors’

t$dᶻil ‘go to bed, animal lies down’ > taadᶻil ‘many go to bed, animals lie down’

As seen in the first two forms, if the first vowel in the base is a schwa, it becomes /aa/ when

reduplicated. Because the long reduplicated vowel attracts stress, this results in the reduction of

the second vowel of the bases in some forms (čaagʷəs ‘wives’, č’aabəs ‘in-laws’), though not in

others (č’aač’as ‘children’, šaadᶻal ‘many go outdoors’). The form siiaʔb ‘noble people’ also

co-exists with the more frequent form siʔiʔab ‘noble people’. This reduplication is also seen

combined with attenuative reduplication in an alternate plural form for ‘girls’, sɬaaɬədəyʔ (from

sɬadəyʔ ‘woman’), with the reduplication of kʷələq ‘other things’ — which is already plural —

126

This form is also recorded as dᶻədᶻadis.

127

This term refers to cross-sex siblings-in-law and a woman’s sister-in-law, but not to a man’s brother-in-law.

333

to form kʷaaləq ‘other people’, and in the word diič’uʔ ‘one person’, the form of the numeral

dəč’uʔ ‘one’ used for counting humans.

Some previous treatments of this reduplicative pattern (Hess & Hilbert 1976; Bates, Hess &

Hilbert 1994) have characterized it as expressing an augmentative rather than a plural; however,

in the present corpus there is only one form which has a clearly augmentative (or, more

accurately, an intensifying) reading:

(415) a. hay dᶻaak’ʷuʔəxʷ əlgʷəʔ tatabtubəxʷ ʔə tiʔiʔiɬ

hay dᶻaak’ʷuʔ=əxʷ əlgʷəʔ tatab–txʷ–b=əxʷ ʔə tiʔiʔiɬ

SCONJ long.time:RDP=now PL speak–ECS–PASS=now PR DIST:PL

‘at long last they were spoken to by them’

(Hess 2006: 79, line 911)

b. … čəxʷa dᶻaak’ʷuʔəxʷ gʷəkʷədxʷ ti haʔɬ biacs

čəxʷa dᶻaak’ʷuʔ=əxʷ gʷə=kʷəd–dxʷ ti haʔɬ biac–s

2SG.COORD long.time:RDP=now SBJ=taken–DC SPEC good meat

‘… finally then you can get its good meat’

(Bates, Hess & Hilbert 1994: 87)

This form, dᶻaak’ʷuʔ ‘at long last, finally’, is derived from the predicate particle dᶻak’ʷuʔ ‘long

(time)’ and expresses intensification (that is, a longer time). Hess & Hilbert (1976: II, 163) also

provide an alternative gloss for č’aač’as ‘children’, ‘still a child’, which might be a

contextualized reading of an augmentative of child (cf. an English expression like ‘big baby’).

These words may be an indication that at one time -V

- reduplication corresponded more

generally to an augmentative meaning, but synchronically this seems to be restricted to only a

few forms.

5.5 Multiple reduplications

In addition to having a great many different reduplicative morphemes, Lushootseed also

allows for the combination of more than one reduplication on a single base; however, only

certain types of multiple reduplications are allowed, and these always occur in a relative fixed

order, as shown in Figure 5:

334

Type II

Type I

-V

Type II

Type III

Type I

Figure 5: Ordering and possible combinations of reduplicands

As indicated in the table, Type II and Type III reduplications can be applied to bases formed by a

previous reduplication. Type II reduplication can combine with Type I reduplication, two of the

special plural patterns, and with bases that have already undergone Type II reduplication. Type

III reduplication expressing diminished effectiveness can only be combined with Type I, and

appears not to combine with any other pattern.

When combined with the Type I distributive reduplication, Type II attenuative reduplication

gives a nominal base meaning ‘N’ the sense of ‘plural small or baby Ns’:

(416) bədaʔ ‘offspring’

bədbədaʔ ‘children’ bibədbədaʔ ‘litter of animals; dolls’

pastəd ‘Caucasian’

paspastəd ‘Caucasians [derogatory]’ papaspastəd ‘Caucasian children’

sgʷəlub ‘pheasant’

sgʷəl’gʷəlub ‘pheasants’ sgʷigʷəl’gʷəlub ‘brood of pheasants’

When applied to verbal bases, the combination of Type I and Type II reduplication creates forms

with plural subjects expressing attenuated states or activities:

(417) q’išəd ‘uncover leg’

128

q’isq’išəd ‘uncover legs’ q’iq’isq’išəd ‘partly uncover legs’

sikʷalus ‘torn eye’ sikʷsikʷalus ‘torn eyes’ siʔsikʷsikʷalus ‘eyes all but torn’

pəkʷ ‘flake off’ pəkʷpəkʷ ‘flake all over’ pipəkʷpəkʷ ‘flake a bit all over’

128

From √q’is ‘be uncovered’ + -šəd ‘lower leg’.

335

Although this combination of reduplicands is transparent, it is not textually frequent and seems

in practice to be largely confined to one or two high-frequency forms (in particular, bibədbədaʔ

referring to the children of anthropomorphic characters in traditional stories).

Type II attenuative reduplication combines to a limited degree with two pluralizing

reduplicative patterns. The first combination covers three forms that combine the Type II

reduplicand with the C

ə- pattern for counting humans, illustrated in (412) above, to make

special forms for counting children:

(418) dəč’uʔ ‘one’ diič’uʔ ‘one person’ didədiič’uʔ ‘one child’

saliʔ ‘two’ səsaʔliʔ ‘two people’ siʔsəsaʔliʔ ‘two children’

ɬixʷ ‘three’ ɬixʷixʷ ‘three people’ ɬiɬxʷixʷ ‘three children’

129

Of these forms, didədiič’uʔ ‘one child’ appears to be based on an unattested C

ə- plural form

*dədiič’uʔ, which of course would be ruled out for semantic reasons; instead, the form is likely

to have been created by analogy with siʔsəsaʔliʔ ‘two children’ and ɬiɬxʷixʷ ‘three children’. It is

not known if there were once more forms of this type in the language.

The second combination consists of Type II plus -V

- pluralization:

(419) stubš ‘man’ stutubš ‘boy’ stuutubš ‘boys’

130

sɬadəyʔ ‘woman’ sɬaɬdəyʔ ‘girl’ sɬaaɬədəyʔ ‘girls’

These are the only two attested forms that combine these two reduplicative patterns.

There are also a number of forms that contain multiple Type II reduplications, the effect of

which is much like that of combining Type I and Type II reduplicands:

(420) bədaʔ ‘offspring’ > bibibədaʔ ‘many young offspring’

č’ƛ’aʔ ‘stone’ > č’ič’iƛ’aʔ ‘many small stone’

qʷɬayʔ ‘log’ > qʷiqʷiqʷɬayʔ ‘many little sticks’

sbadil ‘mountain’ > sbababadil ‘many little mountains’

sgʷəlub ‘pheasant’ > sgʷigʷigʷəlub ‘many fledgling pheasants’

x&aʔx&ayəb ‘giggle’ > x&ax&ax&ax&əybcut ‘many make themselves giggle’

129

Also attested as ɬəɬiʔxʷixʷ.

130

Also stuutububš.

336

With the exception of the final form, all of the textual attestations of iterative Type II

reduplication are on nouns; this last form, x&ax&ax&ax&əybcut ‘many make themselves giggle’ (from

√x&ayəb ‘laugh’) is also exceptional in that it appears to contain three repetitions of the

reduplicand, although the stem x&aʔx&ayəb ‘giggle’ may be lexicalized enough to be considered

the base for the subsequent reduplications. Because of the overlap in meaning of iterative Type II

reduplication with combined Type II + Type I reduplication, forms like those in (420) have been

treated as combinations of the attenuative and the distributive (Hess & Hilbert 1976); however,

the existence of parallel forms such as bibibədaʔ ‘many young offspring’ (Type II + Type II)

and bibədbədaʔ ‘litter of animals; dolls’ (Type II + Type I) makes it more likely that the former

are in fact iterations of the Type II attenuative pattern, and that the iteration of the Type II pattern

has taken on an idiomatic meaning.

Finally, Type III diminished effectiveness reduplication combines with Type I reduplication

to create verbs with the sense of an action performed partially, or ineffectively over and over:

(421) ʔibəš ‘travel’

ʔibʔibəš ‘travel all over’ ʔibibʔibəš ‘meander all over’

ʔuluɬ ‘travel by water’

ʔulʔuluɬ ‘many travel’ ʔululʔuluɬ ‘many out boating’

dᶻalqcut ‘turn around’

dᶻaldᶻalqcut ‘turn repeatedly’ dᶻalaldᶻalqcut ‘turn a bit repeatedly’

131

gʷad ‘speak’

gʷadgʷad ‘talk a lot’ gʷadgʷad ‘converse’

gʷədil ‘sit down’

gʷədgʷədil ‘sit all over’ gʷədədgʷədil ‘just sitting around’

131

This word is used most frequently in the context where it means to look back over one’s shoulder repeatedly.

337

k’ʷalčcut ‘bend self backwards’

k’ʷalk’ʷalčcut ‘bend back repeatedly’ k’ʷalalk’ʷalčcut ‘bend a bit repeatedly’

ƛ’iq ‘emerge’

ƛ’iqƛ’iq ‘emerge all over’ ƛ’iqiqƛ’iq ‘whales emerge all about’

saq’ʷ ‘fly’

saq’ʷsaq’ʷ ‘fly all over’ saq’ʷaq’ʷsaq’ʷ ‘birds wheel in sky’

saxʷəb ‘jump, sprint’

saxʷsaxʷəb ‘run all over’ saxʷsaxʷəb ‘run around for nothing’

xʷəb ‘throw’

xʷəbxʷəb ‘throw repeatedly’ xʷəbəbxʷəb ‘toss from side to side’

This combination of reduplicands seems to be somewhat more common than Type II + Type I.

To the list in (421) we might also add təɬəɬtəɬ ‘be really true’ based on √təɬ ‘be true’ via an

unattested base *təɬtəɬ. There is also at least one example, x&alalx&al ‘serves you right!’, which

appears to be the product of Type III + Type I reduplication of the radical º√x&al ‘get one’s just

desserts’ where the Type III reduplication has the meaning ‘intensive’ rather than ‘diminished

effectiveness’. This radical is poorly attested in the present corpus, however, and the contexts in

which x&alalx&al appear do not make it clear whether it is a productive use of a reduplicated verb

stem or if it constitutes a lexicalized, fixed expression. Similarly, there is a single token in the

textual corpus of Type II + Type III reduplication, čəʔčəlal’dxʷ ‘almost catch something’ (based

on čaldxʷ ‘catch up to someone’ from √čal ‘be overtaken’). This form involves patterns of

vowel reduction and an unusual glottalization of the final /l/ of the radical, and seems not to be a

predictable or productive synchronic form.

338

6 Verbal inflection

6.1 Aspect-marking

One of the two exclusively verbal inflectional categories is aspect, which is marked by a set

of prefixes that appear immediately to the left of the stem, preceding the radical and any

derivational prefixes. The semantics of its aspectual system are at their core fundamentally in-

line with aspectual systems found in a wide variety of the world’s languages, and the language

makes a distinction between three simple aspects — imperfective, perfective, and stative (6.1.1).

However, the description of Lushootseed aspect-marking is complicated by the existence of the

progressive proclitic lə= (Section 3.2), which co-occurs with the aspectual prefixes and interacts

with them morphophonemically and expresses meanings often expressed by aspectual inflections

in other languages. As a result, Lushootseed appears to have two additional “compound aspects”

formed by the combination of the progressive clitic and the perfective and stative prefixes

(6.1.2). Although these have all the trapping of fourth and fifth members of the aspectual

paradigm their semantics and uses in discourse are transparently those of their separate

component parts and from a strictly formal point of view need not be considered as separate

values of the inflectional category of aspect.

6.1.1 Simple aspects

The inflectional category of aspect in Lushootseed has three values — imperfective,

perfective, and stative. Two of the aspects, the perfective and the stative, are clearly marked by

single inflectional prefixes which immediately precede the verb stem, while the third simple

aspect, the imperfective, is a zero:

Ø- ‘imperfective’

ʔu- ‘perfective’

ʔas- ‘stative’

Table 75: Simple aspects

David Beck 10-2-7 2:19 PM

Comment: also goes on question words

339

The meanings of each of these aspects corresponds closely to their meanings in other languages.

The imperfective and perfective aspects in particular seem to conform to Comrie’s (1976:4)

characterizations, the perfective aspect looking “at the situation from outside, without

necessarily distinguishing any of the internal structure of the situation” and the

imperfective aspect looking “at the situation from inside, and as such is crucially

concerned with the internal structure of the situation.” The stative aspect, on the other

hand, treats the situation or event described by the verb (or its outcome) as an established

fact or state of affairs without reference to its beginning or any potential endpoint. The

three aspects are contrasted in (422):

(422) a. haydxʷ əlgʷəʔ stubš tiʔiɬ

Ø–hay–dxʷ əlgʷəʔ stubš tiʔiɬ

IMPF–known–DC PL man DIST

‘they know that he is a man’

[DS Star Child, line 131]

b. tiləb ʔuhaydxʷ ʔuləkʷ’ʷtəb tiʔiɬ ʔalalš ʔə tsi sxʷiyuk’ʷ

tiləb ʔu–hay–dxʷ ʔu–lək’ʷ–t–b tiʔiɬ ʔalalš–s ʔə

immediately PFV–known–DC PFV–eat–ICS–PASS DIST PL–siblings–3PO PR

tsi sxʷiyuk’ʷ

SPEC:FEM Basket.Ogress

‘right away he found out that his siblings had been eaten by the Basket Ogress’

[JS Basket Ogress, line 19]

c. huy ʔəsaydxʷ tašəɬyačiʔbid tiʔiɬ suhuyuds

huy ʔəs–hay–dxʷ tu=ʔas–šəɬ•yačiʔ–bi–d tiʔiɬ

SCONJ STAT–known–DC PAST=STAT–make•hand–MAP–ICS DIST

s=ʔu–huyu–d=s

NM=PFV–made–ICS=3PO

‘he knows how it is done with the hands to make it (shine)’

(Hilbert & Hess 1977: 31)

The sentence in (422a) describes a situation in which two women, having discovered that

an infant is a boy, are preparing to kidnap him. The narrator’s focus here is on the

women’s state of mind (i.e., what they know) at the time of reference and the sentence

340

serves, in narrative terms, as background information motivating the subsequent action of

the characters. In (422b), on the other hand, the use of the perfective aspect places the

focus on the act of discovery (hence the gloss of haydxʷ ‘know’ as ‘find out’), and the

“knowing” event is treated as an atomic whole which includes its beginning point (the

transition from not knowing to knowing). The stative aspect marker in (422c), on the

other hand, portrays the protagonist’s (Mink’s) knowledge as an established state of

affairs at the time of reference. A more detailed discussion of the semantics of aspect in

Lushootseed is somewhat beyond the scope of this grammar; however, some of their

basic uses in narrative, as well as the morphophonemic behaviours of the non-zero

affixes, are detailed in the sections below.

6.1.1.1 Imperfective aspect Ø-

6.1.1.2 Perfective aspect ʔu-

6.1.1.3 Stative aspect ʔas-

The phonologically basic form of the stative aspectual prefix is ʔas-, although the vowel is

unstressed in most word forms and so it most frequently reduced to schwa, giving the form

ʔəs-. The stative prefix also interacts morphophonologically with preceding clitics, as shown

in (423):

341

(423) a. cick’ʷəxʷ ƛ’astagʷəxʷiʔɬəxʷ čəd ʔə tiʔiɬ dbibədbədaʔ

cick’ʷ=əxʷ ƛ’u=ʔas–tagʷəxʷ•iʔɬ=əxʷ čəd ʔə tiʔiɬ

always=now HAB=STAT–hungry•child=now 1SG.SUB PR DIST

d–bi–bəd–bədaʔ

1SG.PO–ATTN–DSTR–offspring

‘my little children usually go hungry’

(Hess 1998: 79, line 48)

b. ɬasxʷəxʷaʔxʷaʔəd čəɬ tiʔiɬ

ɬu=ʔas–xʷəxʷaʔxʷaʔ–d čəɬ tiʔiɬ

IRR=STAT–lightweight–ICS 1PL.SUB DIST

‘we will make that be lightweight’

(Hess 1998: 90, line 81)

c. tastədᶻil ʔal tsiʔiɬ čəgʷas

tu=ʔas–t$dᶻil ʔal tsiʔiɬ č$gʷas–s

PAST=STAT–lie.in.bed at DIST:FEM wife–3PO

‘he was in bed with his wife’

(Hess 2006: 14, line 89)

d. x%ʷul’ basdukʷ tiʔəʔ səsx%qaličtxʷs

x%ʷul’ bə=ʔas–dukʷ tiʔəʔ s=ʔəs–x%q•alič–txʷ=s

only ADD=STAT–anormal PROX NM=STAT–wrapped–ECS=3PO

‘what he had all bundled up was just worthless’

(Hess 1998: 88, line 271)

e. ʔəsʔəx%idəxʷ kʷi ʔaciɬtalbixʷ gʷasʔitutəs

ʔəs–ʔəx%id=əxʷ kʷi ʔaciɬtalbixʷ gʷə=ʔas–ʔitut=əs

STAT–what.happen=now REM people SBJ=STAT–sleep=3SBJ

‘what happened that the people are asleep?’

(Hilbert & Hess 1977: 30)

As these examples show, when preceded by one of the phrasal proclitics (Section 3), the stative

prefix loses its initial glottal stop and causes the elision of the vowel of the proclitic, giving us

the following patterns:

(424) ƛ’u= + ʔas- → ƛ’as-

ɬu= + ʔas- → ɬas-

tu= + ʔas- → tas-

bə= + ʔas- → bas-

gʷə= + ʔas- → gʷas-

132

There is one exception to this pattern in the current corpus where gʷə= + ʔas- → gʷəs–:

342

When strings of proclitics appear before the stative prefix, the clitic adjacent to the aspectual

marker follows the same pattern of coalescence:

(425) xʷiʔəxʷ kʷi stabəxʷ gʷəƛ’ašudxʷ əlgʷəʔ dxʷʔal kʷədiʔ tuk’ʷidəɬdat

xʷiʔ=əxʷ kʷi stab=əxʷ gʷə=ƛ’u=ʔas–šuɬ–dxʷ əlgʷəʔ dxʷʔal kʷədiʔ

NEG=now REM what=now SBJ=HAB=STAT–see–DC PL CNTRPT–at REM.DMA

tu=k’ʷid•əɬ•dat

PAST=how.many•CLS•day

‘they could not see anything for many days’

(Hess 2006: 53, line 263)

In this example, the final /s/ of the stative prefix is lost due to contact with the initial /š/ of the

radical due to a more general phonological, as opposed to specific morphophonological, process.

When the stative prefix follows either of the nominalizing proclitics, s= or dəxʷ=, it simply

loses the initial glottal stop:

(426) a. ɬiltəbəxʷ tsiʔəʔ čəgʷas ʔə tiʔəʔ ɬusəsq’ʷəls

ɬil–t–b=əxʷ tsiʔəʔ čəgʷas–s ʔə tiʔəʔ

give.food–ICS–PASS=now PROX:FEM wife–3PO PR PROX

ɬu=s=ʔəs–q’ʷəl=s

IRR=NM=STAT–cooked=3PO

‘his wife was given it for her to cook’

(Hess 1998: 84, line 166)

b. diɬ cəxʷəsčəbaʔ tiʔəʔ dsəsc’qʷib tsiʔəʔ ʔuʔu

diɬ d=dəxʷ=ʔəs–čəbaʔ tiʔəʔ d=s=ʔəs–c’qʷib

FOC 1SG.PO=ADNM=STAT–backpack PROX 1SG.PO=NM=STAT–get.in.on

tsiʔəʔ ʔuʔu

PROX:FEM little.bit

‘this little bit that I am backpacking is what I got in on’

(Hess 1998: 75, line 249)

(i) wiliq’ʷitəbəxʷ tsiʔəʔ ƛ’aƛ’ac’apəd gʷəsčaləs kʷi x%əčs

wiliq’ʷi–t–b=əxʷ tsiʔəʔ ƛ’aƛ’ac’apəd gʷə=ʔəs–čal=əs kʷi x%əč–s

ask–ICS–PASS=now PROX:FEM ant SBJ=STAT–how=3SBJ REM mind–3PO

‘Ant is asked what is on her mind’

(Hess 1995: 144, line 32)

343

The morphophonemic interaction of the stative prefix with the remaining proclitic, the

progressive, will be discussed in Section 6.1.2 below.

The stative aspect marker also interacts morphophonemically with the propriative prefix bəs-

(Section 2.1.5), losing its final /s/ and triggering elision of the schwa in the following prefix:

(427) a. gʷəl ʔəbsčəgʷas tiʔiɬ təkʷtəkʷəlus ʔə tsiʔiɬ waq’waq’

gʷəl ʔəs–bəs–čəgʷas tiʔiɬ təkʷtəkʷəlus ʔə tsiʔiɬ waq’waq’

SCONJ STAT–PROP–wife DIST owl PR DIST frog

‘Owl had a wife, Frog’

(Hess 2006: 3, line 2)

b. huy čəxʷ tascutəb ʔə tiʔiɬ tabsbədaʔ

huy čəxʷ tu=ʔas–cut–t–b ʔə tiʔiɬ tu=ʔas–bəs–bədaʔ

SCONJ 2SG.SUB PAST=STAT–say–ICS–PASS PR DIST PAST=STAT–PROP–offspring

‘for he who had a daughter told you to’

(Hess 1998: 98, line 203)

The stative aspect is the most frequent aspect used with verbs formed with the propriative prefix

making the ʔəbs- form in (427a) the most common manifestation of the propriative. As shown in

the final word in (427b), the stative prefix can interact simultaneously with morphemes on either

side of it. In this case, interaction with the past proclitic preserves the /a/ vowel of the stative

prefix, which in other respects interacts normally with the propriative prefix, losing its final /s/

and causing the elision of the schwa in the following syllable..

When it precedes the prefix dxʷ(s)- ‘contained’ (Section 2.1.1.1). the stative also loses its

final /s/ and triggers the elision of the following /d/:

(428) a. yəx%i čəd huy ʔəxʷəliʔdubutəb

yəx%i čəd huy ʔəs–dxʷ–həliʔ–dxʷ–but–b

because 1SG.SUB SCONJ STAT–CTD–alive–DC–REFL–PASS

‘because I want to make myself well’

(Hess 1998: 58, line 50)

b. yəx%i ʔəxʷcutəbid ʔə kʷi suhuys

yəx%i ʔəs–dxʷ–cut–əb–bi–d ʔə kʷi sʔuhuys

because STAT–CTD–say–DSD–MAP–ICS PR REM NM=PFV–made=3PO

‘because he thought there was something going on’

(Hess 2006: 14, line 82)

344

c. huygʷasəxʷ tiʔacəc tubšədəd kʷi səxʷcutəbitəbs

huy•gʷas=əxʷ tiʔacəc tubšədəd kʷi s=ʔəs–dxʷ–cut–əb–bi–t–b=s

made•pair=now UNQ Sahaptin REM NM=STAT–CTD–say–DSD–MAP–ICS–PASS=3PO

‘he who is thought to be a Sahaptin is married’

(Hess 1998: 98, line 205)

The most frequent environment for this interaction is where the stative prefix co-occurs with the

dxʷ- prefix when the latter forms a part of the desiderative (Section 2.1.1.7), as in (428a).

Particularly common is the use of the stative with the idiomatic expression dxʷcutəb ‘think’ (lit.

‘want to say’) (428b). (428c) shows the reduction of the morphological sequence s=ʔəs-dxʷ- to

səxʷ-, all that remains of the stative prefix being a schwa.

The frequent use of the stative aspect with dxʷcutəb ‘think’ stems from the fact that the

stative aspect is often used with expressions of thoughts and mental or emotional states. In the

examples in (429), for instance, the stative aspect applies to expressions of mental state:

(429) a. gʷəl ʔəsdᶻaƛ’bid gʷat kʷi sdaʔ ʔə tsiʔəʔ sɬadəyʔ

gʷəl ʔəsdᶻaƛ’bid gʷat kʷi sdaʔ ʔə tsiʔəʔ sɬadəyʔ

SCONJ STAT–confused–MAP–ICS who REM name PROX:FEM PR woman

‘but he was confused about this this woman’s name was’

[HM Star Child, line 117]

b. ʔəsqʷibəxʷ čələp

ʔəs–qʷib=əxʷ čələp

STAT–ready=now 2PL.SUB

‘you guys be ready!’

(Hess 2006: 66, line 588)

Verbs of emotion are also commonly inflected for stative aspect:

(430) a. ʔəsx%iciləxʷ tsiʔəʔ č’ač’as

ʔəs–x%icil=əxʷ tsiʔəʔ č’ač’as

STAT–angry=now PROX:FEM child

‘the girl is angry’

(Hess 1998: 99, line 221)

b. huy cick’ʷəxʷ ʔəsǰuʔiləxʷ

huy cick’ʷ=əxʷ ʔəs–ǰuʔ–il=əxʷ

SCONJ very=now STAT–enjoy–INCH=now

‘she is very happy’

[ML Basket Ogress line 77]

345

c. ʔəx%ʷaq’ʷigʷəd ʔə tiʔəʔ wiw’su

ʔəs–dxʷ–x%ʷaq’ʷ•igʷəd ʔə tiʔəʔ wiw’su

STAT–CTD–troubled•inside.body PR PROX children

‘he was annoyed by the children’

(Hess 1998: 94, line 85)

The stative aspect presents emotions and state of mind as characteristics of the person being

described at the time of reference, as opposed to the other aspects describe the emotional state or

the coming thereinto as an event, often lending a genuine inchoative reading to the clause (i.e.,

‘he got angry’, ‘he becomes angry’, etc.).

Expressions of physical states of being such as illness are also commonly inflected with the

stative aspect:

(431) a. hagʷəxʷ ʔəsx%əɬ

hagʷ=əxʷ ʔəs–x%əɬ

long.time=now STAT–sick

‘he was sick for a long time’

(Hess 1998: 91, line 20)

b. hay, tuʔux%ʷtubəxʷ ʔə kʷi tusʔux%ʷtubs tsiʔəʔ cədiɬ tasq’iƛ’q’iƛ’

hay tu=ʔux%ʷ–txʷ–b=əxʷ ʔə kʷi tu=s=ʔux%ʷ–txʷ–b=s tsiʔəʔ

SCONJ PAST=go–ECS–PASS=now PR REM PAST=NM=go–ECS–PASS=3PO PROX:FEM

cədiɬ tu=ʔas–q’iƛ’–q’iƛ’

s/he PAST=STAT–DSTR–wounded

‘so then this wounded one was taken by those who took her’

(Hess 2006: 20, line 206)

c. yəx%i ƛ’asxʷəlk’ʷəxʷ

yəx%i ƛ’u=ʔas–xʷəlk’ʷ=əxʷ

because HAB=STAT–dizzy=now

‘because she is dizzy’

[ML Basket Ogress, line 83]

The perfective aspect of this type of verb invariably inchoative, lending itself to translations with

English “got” (e.g., ʔux&əɬ ‘he got sick’).

As Hess (1995: 47) also notes, verbs of perception also frequently take the stative aspect:

346

(432) a. x%ʷuləxʷ ʔəšuuc tiʔiɬ ʔay’əds, sčətxʷəd

x%ʷul=əxʷ ʔəs–šuɬ–c tiʔiɬ ʔay’əd–s sčətxʷəd

only=now STAT–see–ALTV DIST companion–3PO bear

‘he just looks at his buddy, Black Bear’

(Hess 1995: 154, line 90)

b. ʔəsluud čəd

ʔəs–lu–d čəd

STAT–hear–ICS 1SG.SUB

‘I hear it’

(Hess 2006: 17, line 143)

c. x%ʷul’ ʔəstkʷadiʔ xʷuʔələʔ ʔal kʷi dəxʷəshuys

x%ʷul’ ʔəs–tkʷadiʔ xʷuʔələʔ ʔal kʷi dəxʷ=ʔəs–huy=s

only STAT–deaf maybe at REM ADNM=STAT–made=3PO

‘he was simply deaf, I guess, to what was done’

(Hess 1998: 97, line 170)

The same verbs with the other other two aspects portray the event of perception in its entirety,

including the inceptive phase of the act (the transition from non-perception to perception),

whereas the stative aspect treats the perception as a mental state of the perceiver, as a fixed state

of affairs at the time of reference.

The stative aspect also naturally lends itself to use with inherently stative radicals such as

those in (433):

(433) a. ʔəsʔitut tiʔiɬ sbiaw

ʔəs–ʔitut tiʔiɬ sbiaw

STAT–sleep DIST coyote

‘Coyote is asleep’

(Hess 1998: 102, line 284)

b. ʔəliqsəxʷ čəd ɬasʔatəbəd

ʔəliqs=əxʷ čəd ɬu=ʔas–ʔatəbəd

even.if=now 1SG.SUB IRR–STAT–dead

‘even if I’ll have died’

(Hess 1998: 91, line 13)

c. ʔəshuygʷəs ʔə təkʷtəkʷəlus ʔə tsiʔəʔ waq’waq’

ʔəs–huy•gʷəs ʔə təkʷtəkʷəlus ʔə tsiʔəʔ waq’waq’

STAT–made•pair PR owl PR PROX:FEM frog

‘Owl and Frog were married’

(Hess 2006: 3, line 7)

347

Stative verbs inflected for the stative aspect present the event as a description of the current

situation or state of affairs. The stative is also used when describing the inherent properties of

persons or objects:

(434) a. huy ʔəsqʷic’

huy ʔəs–qʷic’

SCONJ STAT–lazy

‘as he is (so) lazy’

(Hess 1995: 144, line 51)

b. put ʔəsp’il šqabac tiʔəʔ hikʷ č’ƛ’aʔ

put ʔəs–p’il šqabac tiʔəʔ hikʷ č’ƛ’aʔ

really STAT–flat high•body PROX big stone

‘this big rock is really flat on top’

(Hess 1995: 147, line 4)

It is also commonly found on verbs expressing position or location:

(435) a. x%ʷul’ ʔəxʷpakʷahəb tiʔiɬ sčətxʷəd

x%ʷul’ ʔəsdxʷpakʷahəb tiʔiɬ sčətxʷəd

only STAT–lie.with.rear.up DIST bear

‘Black Bear just lies there with his butt in the air’

133

(Hess 1995: 144, line 36)

b. x%ʷul’ ʔəsq’il ʔal tiʔəʔ q’il’bid

x%ʷul’ ʔəsq’il ʔal tiʔəʔ q’il’bid

only STAT–aboard at PROX canoe

‘it was just on board a canoe’

(Hess 1998: 92, line 38)

c. ʔəscqʷuɬtubəxʷ tiʔəʔ bibaʔəd č’ač’as ʔəskəkiʔiʔɬ

ʔəs–cqʷuɬ–txʷ–b=əxʷ tiʔəʔ bibaʔəd č’ač’as ʔəs–kəkiʔ•iʔɬ

STAT–post–ECS–PASS=now PROX small child STAT–cradleboard•child

‘the small child wrapped on a cradleboard is hung on a post’

(Hess 2006: 40, line 456)

Once again, the use of the stative aspect makes sentences such as these essentially descriptive

statements about a state of affairs extant at the time of reference, as opposed to descriptions of

133

Etymologically-speaking, the verb dxʷpakʷahəb contains the lexical suffix -ah ‘rump’, which is accompanied by

the prefix dxʷ- ‘contained’. Although the prefix continues to interact morphophenomically with the stative marker,

the verb is not synchronically parseable.

348

the protagonist assuming a particular position or a construal of being in the position as an action

or activity.

Because of this essentially descriptive function of the stative aspect, it is frequently used with

environmental expressions, or to set the scene for a new discousrse episode by specifying new or

especially relevant conditions or the presence of particular people or things:

(436) a. ʔəsqʷšaˑb

ʔəs–qʷšab

STAT–foggy

‘it was foggy’

(Hess 2006: 53, line 262)

b. ʔəsk’ʷəɬ ʔal tiʔiɬ cədiɬ x%ɬayʔ dxʷgʷəd

ʔəs–k’ʷəɬ ʔal tiʔiɬ cədiɬ x%ɬayʔ dxʷ–gʷəd

STAT–spill at DIST s/he cedar CNTRPT–down

‘it (blood) is pouring down from that cedar’

(Hess 2006: 18, line 152)

c. ʔəsgʷaadil tiʔacəc ʔaciɬtalbixʷ

ʔəs–gʷaad–il tiʔacəc ʔaciɬtalbixʷ

STAT–down:PL–INCH UNQ people

‘those very people were sitting there’

(Hess 1998: 78, line 20)

134

Indeed, the scene-setting function of the stative means that it is very commonly the aspectused in

opening lines of stories:

(437) a. ʔəsɬaɬlil tiʔiɬ, ʔəsɬaɬlil tiʔiɬ ʔi sgʷəlub ʔi tiʔəʔ qaw’qs

ʔəs–ɬaɬlil tiʔiɬ ʔəs–ɬaɬlil tiʔiɬ ʔi sgʷəlub ʔi tiʔəʔ qaw’qs

STAT–live DIST STAT–live DIST CONJ pheasant CONJ PROX raven

‘they live there, both Pheasant and Raven live there’

(Hess 1998: 78, line 1)

b. ʔəsʔiʔistaʔb tiʔiɬ sbəq’ʷaʔ ʔi tsiʔiɬ x%ʷuʔx%ʷəy’

ʔəs–ʔi–ʔistaʔ–b tiʔiɬ sbəq’ʷaʔ ʔi tsiʔiɬ x%ʷuʔx%ʷəy’

STAT–ATTN–be.like–MD DIST heron CONJ DIST:FEM Little.Diver

‘Heron and Little Diver are like this’

(Hess 2006: 10, line 3)

134

This example is the second half of line 20 in the version of the story in Hess (1998), wich was originally

published with some added grammatical material. It has since been reanalyzed into two lines in accordance with the

the sentence as it is actually spoken on the tape, without the need for any editorial amendments. It is actually line 21

of the story in the lexical database.

349

The first formulation in particular, ʔəsɬaɬlil tiʔiɬ … , is a standard opening to narratives somewhat

on a par with the English “once upon a time … .”

When used with more dynamic radicals, the stative aspect presents the endstate of the event

expressed by the radical as a factual state of affairs that holds at the time of reference, essentially

presenting it as a given rather than as an event or occurrence that transpires during the time the

speaker is focused on:

(438) a. diɬ cəxʷəsbəčalq ʔə tiʔiɬ

diɬ d=dəxʷ=ʔəs–bəč•alq ʔə tiʔiɬ

FOC 1SG.SUB=ADNM=STAT–fallen•game PR DIST

‘that is how I could fell that game’

(Hess 1998: 85, line 209)

b. tiləb ʔugʷəx%agʷil tiʔəʔ tusəsčəbaʔs kʷagʷičəd

tiləb ʔu–gʷəx%–agʷil tiʔəʔ tu=s=ʔəs–čəbaʔ=s kʷagʷičəd

immediately PFV–untied–AUTO PROX PAST=NM=STAT–backpack=3PO elk

‘immediately the elk he had been packing on his back got loose’

(Hess 1998: 87, line 253)

c. diɬəxʷ tiʔiɬ dəxʷəsluutəbs tsiʔiɬ cədiɬ diver, x%ʷuʔx%ʷəyʔ

diɬ=əxʷ tiʔiɬ dəxʷ=ʔəs–lu–t–b=s tsiʔiɬ cədiɬ diver

FOC=now DIST ADNM=STAT–hear–ICS–PASS=3PO DIST:FEM diver

Little.Diver

x,ʷuʔx,ʷ$yʔ

‘that is how Little Diver is heard (i.e., how she sounds)’

(Hess 2006: 13, line 57)

This use of the stative aspect often lends itself to English translations in the perfect aspect:

(439) a. gʷəl ləbəčad tiʔəʔ cədiɬ sxʷiʔxʷiʔs tiʔəʔ səsɬiltəbs

gʷəl lə=bəča–d tiʔəʔ cədiɬ sxʷiʔxʷiʔ–s

SCONJ PROG=fallen–ICS PROX s/he game–3PO

s=ʔəs–ɬil–t–b=s

NM=STAT–give.food–ICS–PASS=3PO

‘and next he sets down the game he has been given’

(Hess 1998: 82, line 124)

350

b. c$ɬul’ ʔ$sqʷib

c$ɬul’ ʔ$–sqʷib

previously STAT–prepared

‘it has been prepared in advance’

(Hess 82, line 132)

c. ʔəsɬiɬič’tub ʔə tiʔəʔ ʔaciɬtalbixʷ

ʔəs–ɬi–ɬič’–txʷ–b ʔə tiʔəʔ ʔaciɬtalbixʷ

STAT–ATTN–cut.up–ICS–PASS PR PROX people

‘these people have it cut up into little pieces’

(Hess 1998: 82, line 133)

This seems consistent with the semantics of both the perfect aspect, which focuses on the state of

affairs resulting from an action or event as it is relevant at the time of reference, and the stative

aspect, which presents an event or its outcome as a state of affairs in effect at the time of

reference.

In addition to being a regular verbal inflection, the stative aspect also appears on question

words, most commonly in questsion that ask about the general situation or extant conditions at

the time of reference:

(440) a. gʷəl ʔəsčal tiʔəʔ sčətxʷəd

gʷəl ʔəs–čal tiʔəʔ sčətxʷəd

SCONJ STAT–how PROX bear

‘and how is it with Black Bear?’

(Hess 1995: 144, line 35)

b. ʔəsʔəx%idəxʷ kʷi ʔaciɬtalbixʷ gʷasʔitutəs

ʔəs–ʔəx%id=əxʷ kʷi ʔaciɬtalbixʷ gʷə=ʔas–ʔitut=əs

STAT–what.happen=now REM people SBJ=STAT–sleep=3SBJ

‘what happened that the people are asleep?’

(Hilbert & Hess 1977: 30)

The sentence in (440a) comes for a context where it is used as a rhetorical device as the narrator

shifts from a description of one of two of the story’s protagonists, Ant, to her foil, Black Bear.

The question in (440b) asks for a reason for the extant state of affairs (that the people are asleep

during their normal waking time, because Raven has mishandled Daylight and made the day too

short). When inflected for the stative aspect, the question word stab ‘what?’ seems to take on a

slightly idiomatic indefinite or nonspecific meaning:

351

(441) a. ʔəstab kʷi gʷədsq’p’ucid

ʔəs–stab kʷi gʷə=d=s=q’p’u–t–sid

STAT–what REM SBJ=1SG.PO=NM=pay–ICS–2SG.OBJ

‘what should I pay you?’

(Hess 2006: 30, line 190)

b. xʷiʔəxʷ tiʔiɬ tuhaʔɬ tul’ʔal tə ʔa tubastab

xʷiʔ=əxʷ tiʔiɬ tu=haʔɬ tul’–ʔal tə ʔa tu=bə=ʔas–stab

NEG=now DIST PAST=good CNTRFG–at NSPEC be.there PAST=ADD=STAT–what

‘whatever was good (to eat) from there was gone’

(Hess 2006: 41, line 478)

In both sentences, stab refers to a set of things that are not clearly defined in context. In (441a),

the speaker (Coyote’s Son, who has been stranded in the Sky World) is completely in the dark as

to what the appropriate payment (to Spider, a supernatural being who has offered assitance)

would be, or (as stated a few lines later) even if he can get his payment to his benefactor. In the

second sentence (441b), stab refers to an undefined and heterogenous class of things (things in

the house that are good to eat).

Another idiomatic, or perhaps fossilized, use of thestative aspect is in certain expressions of

wanting, desiring, or liking based on the nominalized radical sx&aƛ’ ‘desired one’ (Section 8.7):

(442) a. həlaʔb čəxʷ dəsx%aƛ’

həlaʔb čəxʷ d=ʔəs–s–x%aƛ’

really 2SG.SUB 1SG.PO=STAT–NP–desire

‘I really want you’

b. dəsx%aƛ’ kʷi biac

d–ʔəs–s–x%aƛ’ kʷi biac

1SG.PO–STAT–NP–desire REM meat

‘I like meat’

(Bates et al. 1994: 258)

The stative aspect would not normally be expected to appear on a lexical nominalization, and, as

copiously illustrated above, follows rather than precedes the nominalizing clitic s=.

Hess (1995: 46) notes that the stative aspectual marker does not appear on a few verbs that

express intrinsic qualities or properties—specifically, haac ‘long’, luƛ’ ‘old’, and x&ik’ʷ ‘ugly’.

352

However, two of the three verbs on this list are each attested once with stative marking in the

analyzed data currently available:

(443) a. diɬ səsaac ʔə tiʔiɬ sdəxʷiɬ

diɬ s=ʔəs–haac ʔə tiʔiɬ sdəxʷiɬ

FOC NM=STAT–long PR DIST canoe

‘that is the length of a duck-hunting canoe’

(Bates et al 1994: 106)

b. gʷəl kʷatačtub ʔə tudiʔ šəq ʔə tiʔəʔ cədiɬ ʔəsluʔluƛ’ qʷɬayʔ

gʷəl kʷatač–txʷ–b ʔə tudiʔ šəq ʔə tiʔəʔ cədiɬ ʔəs–luʔ–luƛ’

SCONJ climb–ECS–PASS PR DIST.DMA up.high PR PROX s/he STAT–ATTN–old

qʷɬayʔ

cedar

‘and he is taken way up this old cedar tree’

(Hess 2006: 27, line 212)

This probably indicate that, rather than there being a strict constraint against the use use of the

stative with these verbs, the difficulty of finding and eliciting such forms may be of a more

pragmatic nature.

Hess also mentions two other words, hikʷ ‘big’ and haʔɬ ‘good’, that are classified in the

grammar as lexical adverbs (see Table 59) as not taking the stative aspect. In fact, neither of

these would be expected to take any aspectual inflection, although there is one instance in the

corpus of hikʷ ‘big’ with an aspectual prefix on it:

(444) ʔəshigʷəxʷ tiʔəʔ šuƛ’

ʔəs–higʷ=əxʷ tiʔəʔ šuƛ’

STAT–big=now PROX ebb.tide

‘it’s a very low tide now’

(Hess 2006: 12, line 36)

This is the only instance in the corpus of an item from Table 59 taking aspectual marking.

353

ʔəsbuus kʷi tuhuyud əlgʷəʔ

ʔəs–buus kʷi tu=huyu–d əlgʷəʔ

STAT

–four

REM

PAST

=made–

ICS

‘They made four.’

DS Star Child 357

6.1.2 Compound aspects — ləs- and ləcu-

6.2 Passive voice

The passive voice in Lushootseed is formed on transitive stems by the addition of a

morpheme -b ‘passive [PASS]’. The application of the passive suffix results in a form in which

the PATIENT/ENDPOINT (the object of the transitive form) is realized as the subject of the passive

form and the AGENT/INITIATOR (the transitive subject) is realized as an agentive complement

(Section 8.1.7) introduced by the general preposition ʔə:

(445) a. ʔuʔusil ti c’ix%c’ix%

ʔu–ʔusil ti c’ix%c’ix%

PFV–dive SPEC fish.hawk

‘the fish hawk dove’

b. ʔuʔusis ti sʔuladxʷ

ʔu–ʔusil–s–b Ø ti sʔuladxʷ

PFV–dive–ALTV 3SUB SPEC salmon

‘s/he/it dove after the salmon’

c. ʔuʔusisəb ʔə ti c’ix%c’ix% ti sʔuladxʷ

ʔu–ʔusil–s ʔə ti c’ix%c’ix ti sʔuladxʷ

PFV–dive–ALTV–PASS PR DEF fish.hawk DEF salmon

‘the salmon was dived after by the fish hawk’

135

(Hess 1995: 22, ex. 1a–c)

135

Note that the pragmatic uses and discourse functions of the Lushootseed passive are quite different from those of

its English counterpart, and so Lushootseed passives are often more idiomatically glossed as English actives. To

avoid confusion, I have not followed this practice in all cases, even when the results are somewhat stilted English.

354

Sentences with both an overt subject NP and an agentive complement are rare in texts. When

they do occur, the preferred order is that shown in (445c); however, either order of subject and

agentive complement is accepted by speakers and attested in texts:

(446) a. ʔugʷəč’təb ʔə ti č’ač’as ti sqʷəbayʔ

ʔu–gʷəč’–t–b ʔə ti č’ač’as ti sqʷəbayʔ

PFV–search–ICS–PASS PR SPEC child SPEC dog

‘the dog was looked for by the boy’

b. ʔugʷəč’təb ti sqʷəbayʔ ʔə ti č’ač’as

ʔu–gʷəč’–t–b ti sqʷəbayʔ ʔə ti č’ač’as

PFV–search–ICS–PASS SPEC dog PR SPEC child

‘the dog was looked for by the boy’

(Hess 1995: 23, ex. 6a–b)

When the subject of the passive is first- or second-person, it is realized using one of the subject-

markers (Section 8.1.1):

(447) ʔut’uk’ʷtub čəxʷ ʔu ʔə ti luƛ’

ʔu–t’uk’ʷ–txʷ–b čəxʷ ʔu ʔə ti luƛ’

PFV–go.home–ECS–PASS 2SG.SUB INT PR SPEC old

‘were you taken home by the old man?’

(Hess 1995: 24, ex. 10)

In these cases, the relative order of the subject and the agentive complement is invariant and the

subject-marker follows the usual rules for its placement relative to the verb and other ad-verbal

elements (Section 0).

Like the internal causative (Section 2.1.2.1) and the middle (2.1.1.3) suffixes, the passive

shows a certain amount of morphophonemic interaction with its base. Because the vast majority

of transitive stems in Lushootseed are formed (either synchronically or diachronically) from the

combination of a radical and one of the valency-increasing suffixes (2.1.1.7), the

morphophonemics of the passive -b is best described in terms of its interaction with this set of

affixes. Affixation of the passive morpheme to a stem ending in -txʷ or -dxʷ causes the final /xʷ/

of the stem to be realized as /i/:

355

(448) a. ʔuʔəy’dxʷ čəd tsi č’ač’as

ʔu–ʔəy’–dxʷ čəd tsi č’ač’as

PNT–find–DC 1SG.SUB SPEC:FEM child

‘I found the girl’

(Hess 1995: 19)

b. ʔuʔəy’dub čəd ʔə ti č’ač’as

ʔu–ʔəy’–dxʷ–b čəd ʔə ti č’ač’as

PNT–find–DC–PASS 1SG.SUB PR SPEC child

‘I was found by the boy’

(Hess 1995: 33)

With other suffixes, an epenthetic schwa occurs immediately before the passive marker:

(449) a. ʔukʷədatəb ʔə tsi č’ač’as ti sqʷəbayʔ

ʔu–kʷəda–t–b ʔə tsi č’ač’as ti sqʷəbayʔ

PFV–held–ICS–PASS PR SPEC:FEM child SPEC dog

‘the dog was taken hold of by the girl’

(Hess 1995: 22, ex. 5c)

b. ʔuʔux%ʷcəb čəd ʔə ti č’ač’as

ʔu–ʔux%ʷ–c–b čəd ʔə ti č’ač’as

PNT–find–ALTV–PASS 1SG.SUB PR SPEC child

‘I was gone after by the boy’

(Hess 1995: 33)

As noted in Section 2.1.2.1, the passive forms of verbs that exclusively take the [-əš] allomorph

of the internal causative suffix tend to be idiosyncratic. These forms are summarized in Table 76:

ʔišɬš ‘paddle  [canoe]’

ʔišɬtub ‘be paddled’

ɬalš ‘remove from fire’

ɬaltəb ‘be removed from fire’

ɬalšəb ‘be removed from fire’

ƛ’alš ‘wear ’

ƛ’alib ‘be worn’

tagʷš ‘buy ’

tagʷib ‘be bought’

suxʷtəš ‘recognize ’

suxʷtəd ‘be recognized’

Table 76: Passive forms of internal causative stems ending in -š

The passive forms of stems which take either the [-əš] or the [-t]/[-əd] allomorphs of the internal

causative form their passives as [-təb].

In the single instance of the passive form of an inherently transitive, monomorphemic radical

in the current corpus, schwa epenthesis does not occur:

356

(450) a. ʔuɬəgʷəɬ čəɬ ti kikəwič

ʔu–ɬəgʷəɬ čəɬ ti kikəwič

PFV–leave.behind 1PL.SUB SPEC ATTN–hunchback

‘we left Little Hunchback behind’

[LA Basket Ogress, line 121]

b. haˑy ɬəgʷəlb, x%ʷul’ ʔəsq’il ʔal tiʔəʔ q’il’bid

hay ɬəgʷɬ–b x%ʷul’ ʔəs–q’il ʔal tiʔəʔ q’il’bid

SCONJ leave.behind–PASS only STAT–aboard at PROX canoe

‘and then [his corpse] was left, [it] was just aboard his canoe’

(Hess 1998: 92, lines 37–38)

In this example, the final /ɬ/ of the radical √ɬəgʷɬ ‘leave something behind’ becomes /l/ as it

assimilates in voicing to the passive suffix. However, given that this is the only form in which

the passive immediately follows anything other than a valency-increasing suffix, it is impossible

to generalize this process and treat it as anything but an idiosyncratic alternation.

In conservative speech, the passive voice has a special form, -id, used in some subordinate

clauses, as shown in the following examples:

(451) a. ʔəx%id dxʷʔal kʷi gʷəskʷədyitid ʔə tiʔəʔ diʔəʔ gʷəstubšəs

ʔəx%id dxʷ–ʔal kʷi gʷə=s=kʷəd–yi–t–id ʔə tiʔəʔ diʔəʔ

happen CNTRPT–at REM SBJ=NM=held–DAT–ICS–PASS.SBRD PR PROX here

gʷə=stubš=əs

SBJ=man=3SBJ

‘it might happen that he’d be taken from her by them if [it was known] he was male’

[HM Star Child, line 15]

b. gʷəl ckʷaqid ƛ’aslax%dub ʔə tsiʔəʔ luƛ’ tiʔəʔ tucuucid ʔə tsiʔəʔ bədaʔs

gʷəl ckʷaqid ƛ’u–ʔas–lax%–dxʷ–b ʔə tsiʔəʔ luƛ’

then always HAB–STAT–remember–DC–PASS PR PROX:FEM old

tiʔəʔ tu=cut–c–id ʔə tsiʔəʔ bədaʔ–s

PROX PAST=speak–ALTV–PASS.SBRD PR PROX:FEM offspring–3PO

‘and always this old woman would [try to] remember what her daughter would tell her’

[HM Star Child, line 18]

357

c. x%ʷul’ čəd ʔugʷəlaltəb ʔə ti dsluƛ’luƛ’, ʔə tiʔəʔ ʔuč’axʷaptid čəd

x%ʷul’ čəd ʔu–gʷəlal–t–b ʔə ti d–s–luƛ’–luƛ’

only 1SG.SUB PFV–punish–ICS–PASS PR SPEC 1SG.PO–NM–DSTR–old

ʔə tiʔəʔ ʔu–č’axʷ–ap–t–id čəd

PR PROX PFV–clubbed–bottom–ICS–PASS.SBRD 1SG.SUB

‘I just got beaten by my elders, by those who paddled my bottom’

(Hess 2006: 73, line 730)

The -id form of the passive is also used in nominalized clauses serving as predicates:

(452) diɬ ɬudəxʷɬaq’atid kʷi tatačulbixʷ ʔə kʷi gʷat sɬadəyʔ ɬučəbaʔəd

diɬ ɬu=dəxʷ=ɬaq’a–t–id kʷi tatačulbixʷ ʔə kʷi gʷat sɬadəyʔ

FOC IRR=ADNM=set.out–PASS.SBRD REM game.animal PR REM who woman

ɬu=čəbaʔ–d

IRR=laden–ICS

‘what will be laid down is the game animal which will be brought by a woman’

This form of the passive, however, is relatively infrequent even in the speech of the speakers

who still use it, and it is not clear from the present corpus what conditions govern its use — in

fact, judging from the few attestations in the texts, it appears to be in free variation with the -b

form, which is used exclusively by younger speakers in all forms of subordinate clause.

As a final note, it is worth mentioning that the passive voice in Lushootseed does not have

the same restrictions on the person/number of the passive subject as are found in other Salishan

languages:

(453) a. ʔutəlawisəb čəd ʔə ti sqʷəbayʔ

ʔu–təlawil–s–b čəd ʔə ti sqʷəbayʔ

PFV–run–ALTV–PASS 1SG.SUB PR SPEC dog

‘I was run after by the dog’

b. ʔugʷəč’təb čəɬ ʔə ti sqʷəbayʔ

ʔu–gʷəč’–t–b čəɬ ʔə ti sqʷəbayʔ

PFV–search–ICS–PASS 1PL.SUB PR SPEC dog

‘we were looked for by the dog’

c. ʔuʔəy’dub čələp ʔu ʔə tsi č’ač’as

ʔu–ʔəy’dxʷ–b čələp ʔu ʔə tsi č’ač’as

PFV–find–PASS 2PL.SUB INT PR SPEC dog

‘were you guys found by the girl?’

358

d. ʔuʔəƛ’cəb čəxʷ ʔu ʔə ti ʔaciɬtalbixʷ

ʔu–ʔəƛ’–c–b čəxʷ ʔu ʔə ti ʔaciɬtalbixʷ

PFV–come–ALTV–PASS 2SG.SUB INT PR SPEC person

‘were you come for by the people?’

(Hess 1995: 24, ex. 11–14)

However, passives with first- or second-person agentive complements are impossible (Hukari

1976; Jelinek & Demers 1983), although passives with third-person pronouns as agentive

complements are well attested. This issue will be dealt with further in the discussion of agentive

complements in Section 8.1.7.

359

7 Referential expressions

7.1 Deixis

7.2 Possessives

Possessed nouns in Lushootseed are inflected for the person and (in the first- and second-

person) number of their possessor by a rather heterogeneous set of elements which combines

prefixes, suffixes, and (in the first-person plural) a particle from the matrix-subject series

(Section 8.1.1). These are shown in Table 77:

čəɬ

ad-

-ləp

-s

Table 77: Possessive markers

These markers are used both as possessive inflection on nouns and to express the subjects of

nominalized clauses (Sections 7.3 and 8.1.1). When special emphasis is required, pronouns can

co-occur with the first- and second-person possessive markers, as in the following example,

although such sentences seem to be rare:

(

454)

ƛʼubəxʷ x%ʷulʼəxʷ ɬadqʼədᶻax% tsiʔiɬ dəgʷi, pʼuayʼ

ƛʼub=əxʷ x%ʷulʼ=əxʷ ɬu=ad–qʼədᶻax tsiʔiɬ dəgʷi pʼuayʼ

good=now just=now IRR=2SG.PO–entrails DIST:FEM you flounder

‘it’s better that they will just become your entrails, Flounder’

[ML Mink and Tutyika, line 184]

In conservative Lushootseed style, the third-person -s can also co-occur with the following

possessor NP:

(455) x%ʷubts ti hədli

x%ʷubt–s ti hədli

paddle–3PO SPEC Henry

‘Henry’s paddle’

The more frequent pattern found in recorded materials, however, makes use of the preposition ʔə

to introduce the possessor:

360

(456) x%ʷubt ʔə ti hədli

x%ʷubt ʔə ti hədli

paddle PR SPEC Henry

‘Henry’s paddle’

(Hess 1995: 59)

The third-person possessive suffix is not used in these constructions, which follow the obligatory

order Possessed–Possessor. As with the person-markers (Sections 8.1.2 and 8.1.1), possessives in

the third-person do not normally indicate nominal number. When disambiguation is considered

necessary, the pluralizing particle əlgʷəʔ is used:

(457) ʔuʔəy’dxʷ čəd ti sqʷəbayʔs əlgʷeʔ

ʔu–ʔəy’dxʷ čəd ti sqʷəbayʔ–s əlgʷəʔ

PFV–find 1SG.SUB SPEC dog–3PO PL

‘I found their dog’

(Hess 1995: 61, ex. 8b)

The use of this particle is discussed further in Section *.*.

It should be noted that the affixes used to express possession in NPs are homophonous with

the possessive-series of person-markers used for the subjects of nominalized predicate phrases

(Sections 7.4.2 and 8.1.1). For the purposes of this grammar, the two sets of elements are treated

as separate, based primarily on the fact that while possessive affixes in NPs are most frequently

found associated with the phrasal head, even when this is not the first word (other than the

determiner) in the phrase, the same is not true of the possessive-subject markers, which are

(obligatorily) sentence-second clitics:

(458) a. ti hikʷ adpišpiš

ti hikʷ ad–pišpiš

SPEC big 2SG.PO–cat

‘your big cat’

(Hess & Hilbert 1976: I, 131)

b. tux%ʷ čəɬ x%aƛ’txʷ kʷi sƛ’ubləp ʔəsqʷib …

tux%ʷ čəɬ x%aƛ’–txʷ kʷi s=ƛ’ub=ləp ʔəs–qʷib

PTCL 1PL.SUB desire–ECS REM NM=well=2PL.PO STAT–prepared

‘we want you guys to be well-prepared …’

(Hess 2006: 74, line 759)

361

However, as reported in Hess & Hilbert (1976), it is possible, albeit uncommon, for the

possessive affixes to take up other positions in the noun phrase:

(459) a. ti adhikʷ pišpiš

ti ad–hikʷ pišpiš

SPEC 2SG.PO–big cat

‘your big cat’

b. ti adhikʷ adpišpiš

ti ad–hikʷ ad–pišpiš

SPEC 2SG.PO–big 2SG.PO–cat

‘your big cat’

(Hess & Hilbert 1976: I, 131)

While this is clearly clitic-like behaviour (particularly the potential for iteration — see the

discussion of temporal and modal clitics in Section *.*), such constructions are reported by Hess

Hilbert to be rare and are unattested in the present textual corpus. It may be that the possessive

morphemes used in NPs are in transition from being true clitics drawn from the set of person-

markers to being ordinary affixes, though for expository purposes it seems more practical to

distinguish the two sets of morphemes in interlinear glosses to make clear their distinct (though

by no means unrelated — cf. Beck 2000a, 2000b) functions — that of marking possession in NPs

versus that of marking the person and number of subjects in nominalized clauses.

7.3 Modification, attributives, apposition, and coordination

post-posed modifier

huy t’uk’„\x„ ti˙\˙ c\di¬ stubß luÒ’

huy t’uk’„=\x„ ti˙\˙ c\di¬ stubß luÒ’

sconj go.home=now prox he man old

Then this man, the old fellow went home. [sh 109]

mod by both adv and RC

362

gʷaˑʔ ƛ’ulək’ʷəd tiʔəʔ ƛ’uɬčiltub haʔɬ sʔəɬəd

gʷaʔ ƛ’u=ʔu–lək’ʷ–d tiʔəʔ ƛ’u=ɬčil–txʷ–b haʔɬ sʔəɬəd

INTJ

HAB

PFV

–eaten–

ICS

PROX

HAB

=arrive–

ECS

–

PASS

good food

‘But they would eat their good food that would be brought.’

SH line 92

lək’ʷədax%ʷ tiʔəʔ tusɬiltəbs əlgʷəʔ tasq’ʷəl

lək’ʷ–d=ax%ʷ tiʔəʔ tu=s=ɬil–t–b=s əlgʷəʔ

eaten–ics=now prox past=nm=give.food–ics–pass=3po pl

tu=ʔas–q’ʷəl

past=stat–cooked

‘They eat the cooked food they had been given.’

double RC SH40

gʷəl lildəxʷ tiʔiɬ c[əd]iɬ saʔ dəxʷut’uc’s

gʷəl lil–d=əxʷ tiʔiɬ cədiɬ saʔ dəxʷ=ʔu–t’uc’=s

SCONJ

far–

ICS

=now

DIST

s/he bad

ADNM

PFV

–shot=3

‘And they remove these bad things used as shot.’

SH 852

by PP, preposed

b. xʷiʔəxʷ tiʔiɬ tuhaʔɬ tul’ʔal tə ʔa tubastab

xʷiʔ=əxʷ tiʔiɬ tu=haʔɬ tul’–ʔal tə ʔa tu=bə=ʔas–stab

NEG=now DIST PAST=good CNTRFG–at NSPEC be.there PAST=ADD=STAT–what

‘whatever was good from there was gone’

(Hess 2006: 41, line 478)

7.3.1 Modification by adverbs

(460) a. ʔudᶻix%id tiʔəʔ sčəbid ʔal tiʔiɬ luƛ’ qʷɬayʔ

ʔu–dᶻix%i–d tiʔəʔ sčəbid ʔal tiʔiɬ luƛ’ qʷɬayʔ

PFV–break.down–ICS PROX fir.bark at DIST elder tree

‘he broke off and pulled down the bark from an old tree’

(Hess 1998: 71, line 167)

363

(461) huy qa, hikʷ qa biac tiʔiɬ kʷagʷičəd

huy qa hikʷ qa biac tiʔiɬ kʷagʷičəd

SCONJ much big much meat DIST elk

‘well, it was a lot, this elk was really a lot of meat’

(Hess 1998: 84, line 183)

c. tux%ʷ huy hikʷhikʷ haʔɬ sʔuləxʷ stabigʷs …

tux%ʷ huy hikʷ–hikʷ haʔɬ s=ʔu–ləxʷ stabigʷs

just SCONJ DSTR–big good NM=PFV–gathered possession

‘there were lots of good possessions to be gathered’

(Hess 2006: 60, line 437)

7.3.2 Noun–noun attributive constructions

Although less restricted than the use of lexical compounding (Section 2.8), the formation of

noun-noun attributive constructions in Lushootseed is relatively rare, and (unlike English) is

restricted to constructions in which the attributive noun stands in a equative or indexical

semantic relationship to the head noun:

(462) a. tiʔəʔ kiyuuqʷs stətudəq

tiʔəʔ kiyuuqʷs stə–tudəq

PROX seagull PL–slave

‘these seagull-slaves’

(Hess 1995: 91)

b. gʷəl tuləshuyudəxʷ stətudəq tiʔəʔ diʔəʔ səsaʔliʔ təlixʷ biʔbədaʔ tiʔəʔ qaw’qs

gʷəl tu=ləs–huyu–d=əxʷ stə–tudəq tiʔəʔ diʔəʔ sə–saʔliʔ

SCONJ PAST=PROG.STAT–made–ICS=now PL–slave PROX here PL–two

təlixʷ biʔ–bədaʔ tiʔəʔ qaw’qs

blood.brother ATTN–offspring PROX raven

‘and Raven has made slaves of these two small blood brothers’

[HM Star Child, line 98]

c. gʷəɬ x%ʷulč swatixʷəd tiʔəʔ sʔax%ʷuʔ

gʷəɬ x%ʷulč swatixʷəd tiʔəʔ sʔax%ʷuʔ

ASSC sea land PROX clam

‘the clam belongs to the realm of the sea’

(Hess 1998: 33)

In the first two examples here, the attributive noun is equated with its head — thus, in (462a), the

slaves (stətudəq) are seagulls (kiyuuqʷs), while in (462b) the little children (biʔbədaʔ) are blood

brothers (təlixʷ). The attributive construction in (462c) expresses an indexical relationship, the

364

attributive x&ʷulč ‘sea’ indicating which land or realm swatixʷəd is being named. This rather

restricted set of semantic relationships expressed by noun–noun attributive constructions

parallels those associated with nominal predicates (Section 8.3.1); other types of semantic

relationships between nouns are expressed using the associative particle gʷəɬ (Section 7.3.3).

As with other modifiers of nouns, attributive nouns show a certain amount of variable

ordering with respect to their syntactic heads, although this is much more restricted than it is for

words of other classes and it seems to be restricted largely to cases where — for semantic or

morphological reasons — the head of the phrase is unambiguously identifiable:

(463) a. tiʔəʔ ʔiɬt’isu bədaʔs

tiʔəʔ ʔiɬ–t’isu bədaʔ–s

PROX PRTV–younger.relative offspring–3PO

‘his/her youngest son’

(Hess 1995: 91)

b. tiʔəʔ bədaʔs ʔiɬt’isu

tiʔəʔ bədaʔ–s ʔiɬ–t’isu

PROX offspring–3PO PRTV–younger.relative

‘his/her youngest son’

(Hess 1995: 92)

Here, for example, the use of the possessive prefix on bədaʔ ‘offspring’ identifies this noun as

head of the expression, while the partitive prefix ʔiɬ- on t’isu ‘younger relative’ also favours the

interpretation of ʔiɬt’isu as an attributive. Other cases of apparent variation in attributive–head

ordering may in fact reflect an actual reversal of syntactic governance:

(464) a. stubš č’ač’as

stubš č’ač’as

man child

‘boy’ (lit. ‘man-child’)

b. č’ač’as stubš

č’ač’as stubš

child man

‘boy, young man, young male’

(Bates, Hess & Hilbert 1994: 68)

365

In this case, as indicated by the slightly different glosses given the two noun phrases in the

source, the reversal of word-order seems to indicate a change in which of the two nouns is in fact

the head of the construction: in (464a), the expression is headed by č’ač’as ‘child’ and the

attributive classifies the child according to gender, while in (464b), the expression seems to be

headed by stubš ‘man, male’ and č’ač’as ‘child’ tells us about the referent’s age. Given the fact

that noun–noun attributive constructions most often express equative relationship, making the

distinction between, say, kiyuuqʷs stətudəq ‘seagull slaves’ and stətudəq kiyuuqʷs ‘slave

seagulls’ a subtle one, it may be the case that attributive–head ordering is more rigid than it

appears at first glance, and it may be limited to cases such as (463) where morphological clues

help to reduce potential ambiguities.

7.3.3 gʷəɬ ‘associative’

The particle gʷəɬ ‘associative [ASSC]’ takes nominal and bare nominal complements to form

a particular type of restrictive modifying phrase that expresses an association between the

referent of its complement and some other person, place, or thing, the nature of this association

being semantically vague and context-dependent.

136

In many of their uses, gʷəɬ-phrases are

analogous to and interchangeable with the true possessive constructions discussed in Section 7.2

above. In (465), for example, gʷəɬ is used as an expression of a possessor where one might

expect the preposition ʔə. Compare the following examples with, for example, the sentence in

(456):

(465) a. tiʔiɬ q’ilbid gʷəɬ ti dsqa

tiʔiɬ q’ilbid gʷəɬ ti d–sqa

DIST canoe ASSC SPEC 1SG.PO–older.brother

‘my older brother’s canoe’

136

Hess (1998: 33) reports the full form as gʷiɬ , presumably the citation form given by speakers when the word is

elicited out of context; however, this form is unattested in the present corpus and the word is given as gʷəɬ in the

Lushootseed Dictionary, a practice I will follow here.

366

b. gʷəɬ ti dsqa tiʔiɬ q’ilbid

gʷəɬ ti d–sqa tiʔiɬ q’ilbid

ASSC SPEC 1SG.PO–older.brother DIST canoe

‘my older brother’s canoe’

(Hess 1998: 33)

The difference between using gʷəɬ and the preposition ʔə in this type of expression is reported by

Hess & Hilbert (1976: II, 33) to be one of style; however, the two do not appear to be

interchangeable in all constructions (see below). Note also that, unlike a prepositional phrase

headed by ʔə, the gʷəɬ-phrase is variably-ordered relative to the noun it modifies.

137

In this

respect, modifying gʷəɬ-phrases resemble relative clauses (Section 7.4.1).

Another difference with prepositional ʔə-phrases is that the complements of the gʷəɬ-phrases

are frequently bare nouns without determiners

(466) a. gʷ$ɬ b$kʷ stab k’ʷ$luʔ tul$tagʷš

gʷ$ɬ b$kʷ stab k’ʷ$luʔ tul$tagʷš

ASSC all what hide PAST=PROG=bought–ICS

‘it was all kinds of hides that they were buying’

(Hess & Hilbert 1976: II, 42, line 3)

b. swatixʷəd gʷəɬ (ti) ʔaciɬtalbixʷ

swatixʷəd gʷəɬ ti ʔaciɬtalbixʷ

land ASSC SPEC native.people

‘land belonging to (the) native people’

(Hess 1998: 33)

c. tuɬčil dxʷʔal ti swatxʷt$d gʷ$ɬ dxʷsƛ’alb ʔal kʷi tuhaʔkʷ

tu=ɬčil dxʷ–ʔal ti swatxʷt$d gʷ$ɬ dxʷsƛ’alb ʔal kʷi tu=haʔkʷ

PAST=arrive CNTRPT–at SPEC land ASSC Clallam at REM PAST=long.ago

‘they arrived in the land of the Clallam in those days’

(Hess & Hilbert 1976: II, 42, line 2)

d. … ʔal tiʔəʔ diʔəʔ swatixʷəd ʔilucid ʔə tiʔəʔ diʔəʔ gʷəɬ sqaǰət stuləkʷ

ʔal tiʔəʔ diʔəʔ swatixʷəd ʔilucid ʔə tiʔəʔ diʔəʔ gʷəɬ sqaǰət stuləkʷ

at PROX here land river.mouth PR PROX here ASSC Skagit river

‘in the land around the mouth of the Skagit river (lit. ‘this here river of Skagit’)’

[DM Basket Ogress, line 45]

137

See example (474) below for a contextualized example of the less-frequent order shown in (465b).

367

The conditions on the use of the determiner seem to be the same as those at work in other

environments where determiners are optional, depending on the specificity of the referent of the

noun. When the complement of the gʷəɬ-phrase refers to a particular thing or person as opposed

to a generic class of things or general group of people, a determiner is used; otherwise, the

expression is a bare noun phrase. True possessives with ʔə, on the other hand, rarely have bare

nominal complements other than proper names.

Phrases headed by gʷəɬ can also serve as sentence predicates, in which case the noun that is

modified in constructions such as those shown in (465) is expressed as the syntactic subject:

(467) a. huy gʷəɬ sqigʷac tiʔiɬ adsəsčəbaʔ

huy gʷəɬ sqigʷac tiʔiɬ ad=s=ʔəs–čəbaʔ

SCONJ ASSC deer DIST 2SG.SUB=NM=STAT–laden

‘for what you are back-packing is of deer’

[ML Mink and Tutyika I, line 201]

b. hay, gʷəɬəxʷ sqigʷac tiʔiɬ q’ədᶻax%ʷ ʔə tsiʔiɬ p’uay’

hay gʷəɬ=əxʷ sqigʷac tiʔiɬ q’ədᶻax%ʷ ʔə tsiʔiɬ p’uay’

SCONJ ASSC=now deer DIST intestines PR DIST:FEM flounder

‘so, now Flounder’s intestines are (the intestines of) deer’

(lit. ‘so, the intestines of Flounder are of deer’)

[ML Mink and Tutyika I, line 249]

In these examples, the syntactic predicate is the gʷəɬ-phrase gʷəɬ sqigʷac ‘of deer’. This type of

construction can only be formed with gʷəɬ, possessive prepositional phrases headed by ʔə being

unattested as sentence predicates.

Another type of construction that requires gʷəɬ consists of a gʷəɬ-phrase headed by a

determiner. Such expressions have the distribution of ordinary noun-phrases:

(468) saliʔ kʷi gʷəɬ sp’ic’ikʷ

saliʔ kʷi gʷəɬ sp’ic’ikʷ

two REM ASSC Diaper.Child

‘those that belonged to Diaper Child were two’

[DS Star Child, line 356]

368

Again, these syntactic environments are not attested for phrases headed ʔə or any other

preposition. Also unlike prepositional phrases, gʷəɬ-phrases in such constructions, when

modified, are treated as a single unit, as shown in (469):

(469) sax%adəčtəb ʔə tsiʔəʔ čəgʷas tiʔəʔ gʷəɬ sqigʷac ɬusčəɬsəxʷ sɬagʷid

sax%•adəč–t–b ʔə tsiʔəʔ čəgʷas–s tiʔəʔ gʷəɬ sqigʷac

scrape.inside–ICS–PASS PR PROX:FEM wife–3PO PROX ASSC deer

ɬu=s=čəɬ=s=əxʷ sɬagʷid

IRR=NM=make=3PO=now sleeping.mat

‘this of a deer which they’ll make into a sleeping mat was scraped out by his wife’

(Bates, Hess & Hilbert 1994: 64)

In this sentence, the modifying expression, ɬusčəɬsəxʷ sɬagʷid ‘what they will make into a

sleeping mat’ takes scope over the entire construction, gʷəɬ sqigʷac ‘of deer’, rather than simply

over sqigʷac ‘deer’ — that is to say, the sleeping mat is to be made out of a part of the deer,

rather than the deer itself. In this respect, in spite of being demonstrably multi-word syntactic

constructions, gʷəɬ-phrases form tight, coherent semantic units in a way reminiscent of the multi-

word constructions formed with the nominalizing clitics s= and dəxʷ= (Section 7.4.2).

In terms of its meanings, gʷəɬ is rather diffuse. As noted above, one of its most frequent

functions is to express something akin to possession, leading to the gloss of gʷəɬ in some sources

as ‘belonging to’ (e.g., Hess & Hilbert 1976; Hess 1998):

(470) a. diɬ gʷəɬ tiʔiɬ ʔi wiɬwiɬ ʔi sʔadad

diɬ gʷəɬ tiʔiɬ ʔi wiɬwiɬ ʔi sʔadad

FOC ASSC DIST and snipe and magpie

‘it (the weir) belonged to Snipe and Magpie’

(Hess 2006: 37, line 376)

b. xʷiʔəxʷ kʷi bəstab gʷətugʷəɬ tiʔiɬ sč’istxʷs tusəsx%ədyids

xʷiʔ=əxʷ kʷi bə=stab gʷə=tu=gʷəɬ tiʔiɬ sč’istxʷ–s

NEG=now REM ADD=what SBJ=PAST=ASSC DIST husband–3PO

tu=s=ʔəs–x%əd–yi–d=s

PAST=NM=STAT–push.aside–DAT–ICS=3PO

‘there was again nothing of what would have been her husband’s, whom it had been set

aside for’

(Hess 2006: 43, line 45)

369

In both of these examples, the choice of gʷəɬ over a plain possessive expression using ʔə seems

to be syntactically-conditioned. (470a) shows a gʷəɬ-phrase used as a syntactic predicate, a role

that, as noted above, seems not to be open to prepositional phrase headed by ʔə. The gʷəɬ-phrase

in (470b) acts as a modifier of stab ‘what’, indicating that the non-existent items (food provided

for a family by the woman’s brothers) belongs to the husband. This line occurs in a context

where the food that was to have been shared by the family (a portion of which would have gone

to the husband), is divided up between the woman and her children, leaving nothing for the

husband. Thus, the husband’s portion of the food is hypothetical, necessitating the use of the

subjunctive proclitic gʷə=. In the indicative mood, the same expression might have been formed

with ʔə (i.e., stab ʔə tiʔiɬ sč’istxʷs ‘her husband’s’); however, the preposition ʔə is not a

legitimate host for the subjunctive proclitic.

A similar phenomenon occurs in (471), where the use of gʷəɬ to express a type of possession

is required by the presence of a true possessor of the same item:

(471) diɬ sdaʔs, gʷəɬ ʔaciɬtalbixʷ, həlaʔb sdaʔs tsiʔiɬ k’aʔk’aʔ

diɬ s–daʔ–s gʷəɬ ʔaciɬtalbixʷ həlaʔb s–daʔ–s tsiʔiɬ k’aʔk’aʔ

FOC NP–name–3PO ASSC native.people really NP–name–3PO DIST:FEM crow

‘that is her native-people’s name, the true name of Crow’

(Hess 1998: 57, line 23)

Here, the noun sdaʔ ‘name’ (derived from the verbal radical √daʔ ‘be named’) is marked for a

third person possessor by the suffix -s, which refers back to Crow, the protagonist of the

preceding discourse and the more immediate possessor of the name. At the same time, the

speaker wishes to make it clear that this particular name of Crow is the native people’s name for

her, and so gʷəɬ is used to express the native people’s (less immediate) “possession” of the

name.

138

Another possibility here is that there is a contrast between the use of the plain possessive to express the actant

corresponding to the subject of the nominalized verb (in this case Crow, the one who is named) and the use of gʷiɬ

to express an entity that is not part of the core valency of the radical. Given the textual infrequency of gʷiɬ in the

present corpus, however, this will have to be left to the realm of speculation; even with a considerably larger corpus,

370

The use of gʷəɬ in (471) is also reminiscent of another frequent use of the particle in the

specification of types or genera of items:

(472) a. tiʔəʔ k’ʷəluʔ gʷəɬ tatačulbixʷ, k’ʷəluʔ, diɬ shududs əlgʷəʔ

tiʔəʔ k’ʷəluʔ gʷəɬ tatačulbixʷ k’ʷəluʔ diɬ s=hudu–d=s əlgʷəʔ

PROX hide ASSC game.animal hide FOC NM=burn–ICS=3PO PL

‘the hides of animals, hides, that is what they burned’

[DS Star Child, line 330]

b. tsiʔəʔ ƛ’ugreen tə ƛ’ustababacs waw’lis ʔi tsiʔəʔ diʔəʔ ƛ’uk’ʷak’ʷt’ad x%əɬ ti

ƛ’ux%ʷiqʷac gʷəɬ tiʔəʔ woods

tsiʔəʔ ƛ’u=green tə ƛ’u=s=tab•abac=s wa–w’lis ʔi tsiʔəʔ

PROX:FEM HAB=green NSPEC HAB=NM=do•body=3PO ATTN–frog and PROX:FEM

diʔəʔ ƛ’u=k’ʷa–k’ʷt’ad x%əɬ ti ƛ’u=x%ʷiqʷac gʷəɬ tiʔəʔ woods

here HAB=ATTN–mouse seemingly HAB=light.green ASSC PROX woods

‘Little Frog and Little Mouse’s body-covering is this green like the pale green of the

woods’

[MW Star Child, line 79]

Similarly, in (473), gʷəɬ is used to specify which particular realm, of the various mystical realms

in Lushootseed cosmology, the item in question (sʔax&ʷuʔ ‘clam’) belongs to:

(473) gʷəɬ x%ʷulč swatixʷəd tiʔəʔ sʔax%ʷuʔ

gʷəɬ x%ʷulč swatixʷəd tiʔəʔ sʔax%ʷuʔ

ASSC sea land PROX clam

‘the clam belongs to the realm of the sea’

(Hess 1998: 33)

Expressions such as these are generally amenable to translation with the English preposition of

or, particularly in cases such as (472a) and (473) with bare nominal complements, as compounds

(e.g., animal hide, sea realm).

Another use of gʷəɬ is to indicate affiliation or geographical point of origin:

it seems like final confirmation of this hypothesis would require extensive work with native speaker consultants who

could verify the unacceptability of particular gʷiɬ-forms, whose absence from the corpus might be simply stylistic or

accidental.

371

(474) a. baˑlgʷas buʔqʷ kʷi bək’ʷ sčads ʔə ti buʔqʷ gʷəɬ x%ʷəlč, tiʔiɬ bəgʷəɬ t’aq’t, gʷəɬ spaɬx%ad

buʔqʷ

balgʷas buʔqʷ kʷi bək’ʷ s=čad=s ʔə ti buʔqʷ gʷəɬ

all.kinds waterfowl REM all NM=where=3PO PR SPEC waterfowl ASSC

x%ʷəlč tiʔiɬ bə=gʷəɬ t’aq’t gʷəɬ spaɬx%ad buʔqʷ

sea DIST ADD=ASSC inland ASSC tidal.flats waterfowl

‘there were all kinds of Duck People (waterfowl) that came from everywhere, Duck

People from the sea and those from inland, Duck People from the tidal flats’

(Hess 2006: 63, line 457)

b. huy stabigʷs ʔal tiʔəʔ gʷəɬ dibəɬ tiʔiɬ sʔuləx% ʔal tiʔəʔ gʷəɬ tiʔəʔ dəxʷʔas əlgʷəʔ

huy stabigʷs ʔal tiʔəʔ gʷəɬ dibəɬ tiʔiɬ sʔuləx% ʔal tiʔəʔ gʷəɬ

SCONJ possession at PROX ASSC we DIST dentalia at PROX ASSC

tiʔəʔ dəxʷ=ʔa=s əlgʷəʔ

PROX ADNM=be.there=3PO PL

‘but the dentalia in that place they come from are valuable when they are in our place’

(lit. ‘but the dentalia at the (place) of where they are are (valuable) possessions at ours’)

(Hess 2006: 66, line 576)

The example in (474b) also illustrates another use of gʷəɬ with personal pronouns as

complements, forming expressions that might be loosely translated as possessive pronouns:

(475) a. gʷəɬ dibəɬ b. gʷəɬ gʷat c. gʷəɬ stab

gʷəɬ dibəɬ gʷəɬ gʷat gʷəɬ stab

ASSC we ASSC who ASSC what

‘ours’ ‘whose’ ‘its/something’s’

(Hess 1998: 33)

Although the glosses given here lead to the impression that such expressions are in competition

with the true possessive pronouns (Table 69), it seems that the latter are restricted to genuine

possessive contexts, whereas the gʷəɬ expressions appear when the relationship being expressed

is more general than those covered by possessives.

The attributive particle can also be used to express other possessive-like relations, such as

that between a story and that story’s protagonist:

(476) diɬ tušac’ ʔə tiʔiɬ gʷəɬ sgʷəlub tiʔiɬ ʔi tiʔiɬ qaw’qs

diɬ tu=s–šac’ ʔə tiʔiɬ gʷəɬ sgʷəlub tiʔiɬ ʔi tiʔiɬ qaw’qs

FOC PAST=NP–end PR DIST ASSC pheasant DIST and DIST raven

‘that is the end of the (story) of Pheasant, (of) him and Raven’

(Hess 1998: 89, line 310)

372

Clearly, the main character of a story resembles a possessor in the sense that it is a reference

point for identifying which particular story is being referred to, but there is also a sense in which

the semantic relation between the story and its protagonist falls outside the ordinary uses of the

possessive construction.

139

The story is “about” Pheasant rather than “belonging to” him.

Similarly, in the example in (477), gʷəɬ expresses a loose associative relationship between

raccoons (x&aʔx&alus) and the fact that they must be butchered, used to identify a particular point

in time:

(477) ʔalil tiʔəʔ gʷəɬ x%aʔx%alus

ʔalil tiʔəʔ gʷəɬ x%aʔx%alus

come.time PROX ASSC raccoon

‘when it came time to butcher the raccoons’

(lit. ‘when it comes to this of the raccoons’)

(Bates, Hess & Hilbert 1994: 101)

The interpretation of this sentence depends crucially on the addressee’s ability to make the

associative link between the raccoons and their knowledge that, having been killed, they will be

butchered; in another context, presumably, the same expression might be used to refer to the

hides of the raccoons (cf. 471a), their intestines (467), or (perhaps in a story) their land or

territory (466b). This seems typical of the uses of the attributive particle, which covers a broader

and somewhat more diffuse semantic range than the true possessive construction. Consider

another example, in which gʷəɬ is used to identify an unnamed object (the fletching on an arrow)

through reference to an artifact with which it is associated (q’čic ‘bow’):

139

However, it is possible to express the same idea using an ordinary possessive construction:

(i) hay, ʔa tiʔəʔ syəyəhub ʔə tiʔiɬ sčətxʷəd ʔi tsiʔiɬ ƛ’aƛ’ac’apəd

hay ʔa tiʔəʔ syəyəhub ʔə tiʔiɬ sčətxʷəd ʔi tsiʔiɬ ƛ’aƛ’ac’apəd

SCONJ be.there PROX legend PR DIST bear and DIST:FEM ant

‘so, there is a traditional story about Black Bear and Ant’

(Hess 1995: 143, line 1)

The difference appears to be one of style (cf. English the story of Bear and Ant vs. the story about Bear and Ant).

373

(478) gʷəl diɬ xʷuʔələʔ ƛ’ušəɬs x%əɬ ti tiʔiɬ gʷəɬ q’čic gʷədəxʷut’uc’il, dəxʷuxʷiʔxʷiʔ ʔə tiʔəʔ

bədaʔs

gʷəl diɬ xʷuʔələʔ ƛ’u=s=šəɬ=s x%əɬ ti tiʔiɬ gʷəɬ q’čic

SCONJ FOC maybe HAB=NM=make=3PO seemingly DIST ASSC bow

gʷə=dəxʷ=ʔu–t’uc’il dəxʷ=ʔu–xʷiʔxʷiʔ ʔə tiʔəʔ bədaʔ–s

SBJ=ADNM=PFV–shoot ADNM=PFV–hunt PR PROX offspring–3PO

‘and (those red feathers), I guess, are what (Coyotes’s) son apparently uses to make

(fletching) for the bow he

would shoot with, for his hunting’

(lit. ‘and that, I guess, is what he

apparently would make that of the bow he

would shoot

with, for (Coyotes’s) son

’s hunting’)

(Hess 2006: 25, line 61)

Note that in this particular instance, the definitive relationship is not between the fletching and

the arrow (t’isəd), but between the fletching and the bow — a relationship that is much more

broadly associative rather than the specific, immediate relationship between part and whole or

possessor and possessed. Likewise, in the following examples, gʷəɬ is combined with sɬadəyʔ

‘woman’ (479a) and stubš ‘man’ (479b) to form expressions that might best be translated as

‘used for women’ and ‘used for men’, respectively:

(479) a. yəx%i kʷi tuhaʔkʷ tuʔaciɬtalbixʷ ʔa kʷi gʷəɬ sɬadəyʔ saʔliʔɬ gʷəsɬadəyʔəs kʷi adbədaʔ

yəx%i kʷi tu=haʔkʷ tu=ʔaciɬtalbixʷ ʔa kʷi gʷəɬ sɬadəyʔ

because REM PAST=long.time PAST=people be.there REM ASSC woman

s–haʔl•iʔɬ gʷə=sɬadəyʔ=əs kʷi ad–bədaʔ

˹NP–stop.crying•child˺ SBJ=woman=3SBJ REM 2SG.PO=offspring

‘because long ago people had child-comforting (speech) for girls if your child is a girl’

[DS Star Child, line 97]

b. gʷəɬ stubš saʔliʔɬ tiʔiɬ ƛ’usaʔliʔɬs

gʷəɬ stubš s–haʔl•iʔɬ tiʔiɬ ƛ’u=s=haʔl•iʔɬ=s

ASSC man ˹NP–stop.crying•child˺ DIST HAB=NM=stop.crying•child=3PO

‘her comforting of the child was the child-comforting (speech) for boys’

[DS Star Child, line 124]

Here, the noun saʔliʔɬ ‘child-comforting speech’ is modified by a gʷəɬ-phrase which indicates

which gender, male or female, the style of speech is appropriate for (reflecting the Lushootseed

custom of care-givers using gender-specific expressions when caring for young children). As in

the previous examples, the semantic relation between sɬadəyʔ or stubš and saʔliʔɬ is rather loose

and contextually defined, resembling more than anything else the semantic relationship between

374

nouns in English noun-noun compounds. This relationship is either idiosyncratic or lexicalized

for a particular pair of nouns (e.g., post office, blockhead) or is determined pragmatically by

context and the semantic affordances of the component words (e.g., alligator shoes ‘shoes made

from alligator skin’, ‘shoes worn by an alligator’, ‘shoes shaped like alligators’, etc., but not

‘shoes used to construct alligators’ [cf. floor tiles] or ‘shoes eaten by alligators’ [cf. rabbit

pellets]). The variety of semantic relations between the parts of the latter, non-lexicalized type of

compound bears a close resemblance to the variety of associative relations mediated by gʷəɬ,

suggesting that this particle is, in effect, a way of doing the job of noun-noun compounding in a

language that otherwise, as noted in Section 2.8 above, does not have a productive system of

genuine compound formation and a restricted system of noun–noun attributive constructions

(Section 7.3.2).

7.3.4 Appositive phrases

Lushootseed noun phrases also allow for modification by appositive noun phrases, such as

those shown in (480)

(480) a. tiʔəʔ tusč’istxʷ, tuq’iyaƛ’əd

tiʔəʔ tu=sč’istxʷ tu=q’iyaƛ’əd

PROX PAST=husband PAST=slug

‘her former husband, the late Slug’

(Hess 1995: 92)

b. ʔa tsiʔiɬ qiʔqəladiʔ, bədaʔs

ʔa tsiʔiɬ qiʔqəladiʔ bədaʔ–s

be.there DIST:FEM Qiqeladi offspring–3PO

‘there was Qiqeladi, her daughter’

[ML Basket Ogress, line 22]

Appositive nouns invariably follow the head of the noun phrase that contains them and are

generally separated from it by a pause or some other prosodic boundary. In a few cases, a single

noun phrase may contain more than one appositive:

375

(481) cuucəxʷ tiʔəʔ č’ač’as, bibədaʔs, stutubš

cut–c=əxʷ tiʔəʔ č’ač’as bi–bədaʔs stu–tubš

say–ALTV=now PROX child ATTN–offspring–3PO ATTN–man

‘she said to her child, her little son, the little man’

[AW Basket Ogress, line 78]

As seen in these examples, appositives are generally bare nouns, although there are a few

examples in the corpus of appositives introduced by determiners:

(482) a. tsiʔəʔ čəgʷas, tsiʔəʔ x%ʷuʔx%ʷəy’

tsiʔəʔ čəgʷas–s tsiʔəʔ x%ʷuʔx%ʷəy’

PROX:FEM wife–3PO PROX:FEM helldiver

‘his wife, Helldiver’

(Hess 1995: 92)

b. yəx%i tuʔəbsčəgʷasəxʷ ʔə tsiʔəʔ diʔəʔ, tsiʔəʔ ƛ’uwaw’əq’wəq’ kʷsi tučəgʷas

yəx%i tu=ʔəs–bəs–čəgʷas=əxʷ ʔə tsiʔəʔ diʔəʔ tsiʔəʔ

because PAST=STAT–PROP–wife=now PR PROX:FEM here PROX:FEM

ƛ’u=waw’əq’wəq’ kʷsi tu=čəgʷas–s

HAB=Little.Green.Frog REM:FEM PAST=wife–3PO

‘because he had as a wife this one, this Little Green Frog [who was] his wife’

[HM Star Child, line 183]

All the cases in the corpus involve the apposition of proper names; however, proper names can

also be used appositively without determiners:

(483) a. huy, puspusutəbəxʷ ʔə tiʔəʔ diʔəʔ hikʷ sɬadəyʔ, ʔax%ʷadus

huy pus–pusu–t–b=əxʷ ʔə tiʔəʔ diʔəʔ hikʷ sɬadəyʔ ʔax%ʷadus

SCONJ DSTR–throw–ICS–PASS=now PR PROX here big woman Basket.Ogress

‘well then, he was thrown at over and over by the big woman, Basket Ogress’

[DM Basket Ogress, line 38]

b. puˑtəxʷ ʔəsɬuqʷač tiʔiɬ bibščəb ʔi tiʔiɬ suʔsuqʷaʔs, tətyika

put=əxʷ ʔəs–ɬuqʷ•ač tiʔiɬ bi–bščəb ʔi tiʔiɬ suʔ–suqʷaʔ–s

really=now STAT–peeled•head DIST ATTN–mink and DIST ATTN–younger.sibling

tətyika

Tutyika

‘Young Mink and his little younger brother, Tutyika, were completely bald’

(Hess 1995: 142, line 45)

Given that, in general, appositive phrases are not used referentially to identify individuals so

much as to qualify or reinforce the identity of an individual (the referent of the noun phrase in

376

which it is contained), it seems likely that the principles regulating the (non)-use of determiners

with appositives is the same as the principles regulating the use of bare noun phrases discussed in

Section 7.1). As far as the use of determiners with proper nouns is concerned, it seems likely that

the distinction between expressions such as those in (482) vs. those in (483) has to do with

whether or not the proper name is being used as a link between the referent of the larger noun

phrase and some person previously named in discourse (in which case a determiner would be

required), or whether the proper noun is simply being supplied as additional information about

the referent.

Even though appositives tend not to have determiners, they are by no means always simple

bare NPs; not infrequently, appositive nouns are possessed (as, for example, in 480b above) and

can have their own modifiers, as in (484):

(484) a. tiʔiɬ taʔtəmi, haʔɬ sč’ač’as čəɬ

tiʔiɬ taʔ–təmi haʔɬ s–č’ač’as čəɬ

DIST ATTN–Tommy good NP–child 1PL.PO

‘little Tommy, our good youngster’

(Hess 1995: 92)

b. xʷəbtəbaxʷ ʔəsq’ʷuʔ ʔə tiʔacəc bibədaʔs, miʔman’ səsbədaʔəbs tsiʔiɬ cədiɬ waq’waq’

xʷəb–t–b=axʷ ʔəs–q’ʷuʔ ʔə tiʔacəc bi–bədaʔ–s

throw.away–ICS–PASS=now STAT–gather PR UNQ ATTN–offspring–3PO

miʔman’ s=ʔəs–bədaʔ–b=s tsiʔiɬ cədiɬ waq’waq’

small NM=STAT–offspring–MD=3PO DIST:FEM he frog

‘Frog was thrown away together with her little baby, the little (one) that she had borne’

(Hess 2006: 8, line 124)

377

c. ʔaˑhəxʷ ʔal tiʔiɬ dəxʷəsciq’itəbs ʔal tiʔiɬ hud dxʷʔal tusʔatəbəds tsi sxʷəyuq’ʷ,

sƛ’alqəb dxʷʔal kʷi wiw’su

ʔa=əxʷ ʔal tiʔiɬ dəxʷ=ʔəs–ciq’i–t–b=s ʔal tiʔiɬ hud

be.there=now at DIST ADNM=STAT–poke–ICS–PASS=3PO at PROX fire

dxʷ–ʔal tu=s=ʔatəbəd=s tsi sxʷəyuq’ʷ sƛ’alqəb dxʷ–ʔal

CNTRPT–at PAST=NM=die=3PO SPEC:FEM Basket.Ogress monster CNTRPT–at

kʷi wiw’su

REM children

‘there was Basket Ogress, monster to the children, where she had been poked into the

fire until she died’

[ML Basket Ogress, line 109]

The appositive noun in (484a), sč’ač’as ‘youngster’, is modified by haʔɬ ‘good’, while in (484b)

the appositive miʔman’ ‘small’ (used here as a metonymic expression for a small person) is

modified by an s=nominal (Section 7.4.2.1) formed on the verb phrase ʔəsbədaʔəb ‘she has

given birth’. The final example, (484c), shows an appositive noun, sƛ’alqəb ‘monster’, modified

by a following prepositional phrase, dxʷʔal kʷi wiw’su ‘for (lit. ‘towards’) the children’. In

general, the ordering of modifying elements in the appositive phrase seems identical to the

ordering of modifiers in noun phrases in general.

7.3.5 Coordination

Coordinate noun phrases are formed using the conjunction ʔi ‘and’,

140

which is used to

coordinate two full noun phrases, as in (485):

(485) a. tuhuyucut tiʔəʔ qaw’qs ʔi tiʔəʔ bibščəb

tu=huyu–cut tiʔəʔ qaw’qs ʔi tiʔəʔ bi–bščəb

PAST=made–REFL PROX raven and PROX ATTN–mink

‘Raven and Little Mink prepared themselves’

(Hess 1995: 92)

b. ʔəsɬaɬlil tiʔiɬ ʔəsɬaɬlil tiʔiɬ sgʷəlub ʔi tiʔəʔ qaw’qs

ʔəs–ɬaɬlil tiʔiɬ ʔəs–ɬaɬlil tiʔiɬ sgʷəlub ʔi tiʔəʔ qaw’qs

STAT–live DIST STAT–live DIST pheasant and PROX raven

‘they dwelled (there), Pheasant and Raven dwelled (there)’

(Hess 1998: 78, line 1)

140

Southern Lushootseed uses yəxʷ rather than ʔ i.

378

c. ɬacəxʷɬčiltxʷ dxʷʔal kʷsi adčəgʷas ʔi kʷi adbədbədaʔ

ɬu=ad=dəxʷ=ɬčil–txʷ dxʷ–ʔal kʷsi ad–čəgʷas ʔi kʷi

IRR=2SG.PO=ADNM=arrive–ECS CNTRPT–at REM:FEM 2SG.PO–wife and REM

ad–bəd–bədaʔ

2SG.PO–DSTR–offspring

‘so that you can bring it to your wife and children’

(Hess 1998: 86, line 237)

While coordinated NPs with two members are the most frequent, longer lists can be formed with

ʔi, which is repeated after every item in the list:

(486) ƛ’al’ basʔistaʔ dxʷʔal tiʔəʔ təbɬ ʔi tiʔiɬ dəxʷudxʷliq’ʷusəbs ʔi kʷədiʔ ƛ’udəxʷəsx%əqšads

ʔi kʷədiʔ ƛ’udəxʷəsx%əqlax%ads ƛ’udəxʷǰuʔadads

ƛ’al’ bə=as–ʔistaʔ dxʷ–ʔal tiʔəʔ təbɬ

also ADD=STAT–be.like CNTRPT–at PROX ochre

ʔi tiʔiɬ dəxʷ=ʔu–dxʷ–liq’ʷ•us–əb=s

and DIST ADNM=PFV–CTD–paint•face–MD=3PO

ʔi kʷədiʔ ƛ’u=dəxʷ=ʔəs–x%əq•šad=s

and REM.DMA HAB=ADNM=STAT–wrapped•leg=3PO

ʔi kʷədiʔ ƛ’u=dəxʷ=ʔəs–x%əq•l•ax%ad=s

and REM.DMA HAB=ADNM=STAT–wrapped•CNN•arm=3PO

ƛ’u=dəxʷ=ǰuʔadad=s

HAB=ADNM=regalia=3PO

‘it is also the same for the ochre and that used for painting one’s face and that used for

wrapping the legs and that used for wrapping the arms, used in ceremonial regalia’

(Hess 1998: 92, line 32)

In terms of their distribution, coordinated NPs are identical to ordinary noun phrases and can be

the arguments of verbs, as in (485a) and (485b), the complements of prepositions, (485c) and

(486), and possessors of other nouns, as in (487):

(487) hay, ʔa tiʔəʔ syəyəhub ʔə tiʔiɬ sčətxʷəd ʔi tsiʔiɬ ƛ’aƛ’ac’apəd

hay ʔa tiʔəʔ syəyəhub ʔə tiʔiɬ sčətxʷəd ʔi tsiʔiɬ ƛ’aƛ’ac’apəd

SCONJ be.there PROX legend PR DIST bear and DIST:FEM ant

‘so, there is the legend of Black Bear and Ant’

(Hess 1995: 143, line 1)

Coordinated appositive phrases (Section 7.3.4) are also attested:

379

(488) kʷədubəxʷ ʔə ti tubəʔalalšs, swuqʷad ʔi tiʔiɬ x%əwawq’

kʷəd–dxʷ–b=əxʷ ʔə ti tu=bə=ʔal–alš–s swuqʷad ʔi tiʔiɬ

taken–DC–PASS=now PR SPEC PAST=ADD=PL–sibling–3PO loon and DIST

x%əwawq’

Big.Diver

‘both her brothers, Loon and Big Diver, got her’

(Hess 2006: 20, line 205)

As is typical of appositive phrases, the initial determiner introducing swuqʷad ‘loon’ is omitted

here, although the second member of the coordinated NP maintains its determiner.

Coordinated NPs are also attested as syntactic predicates:

(489) a. swuq’ʷadiʔ ʔi x%ʷtis ʔi stab

swuq’ʷadiʔ ʔi x%ʷtis ʔi stab

loon and silver.diver and what

‘there were loons and silver divers and other kinds’

(Hess 2006: 75, line 798)

b. tiʔəʔəxʷ sɬukʷalb ʔi tiʔəʔ ɬukʷaɬ

tiʔəʔ=əxʷ sɬukʷalb ʔi tiʔəʔ ɬukʷaɬ

PROX=now moon and PROX sun

‘(they) now are the moon and the sun’

[MW Star Child, line 108]

As with the coordinated appositive phrases in (488), the principles governing the use/non-use of

determiners are the same as those governing the use of determiners in other types of nominal

predicate (Section 8.3.1).

In addition to coordinating full NPs, ʔi can also coordinate pronominal elements with NPs, as

in (490), which shows the coordination of a pronominally-used determiner, tiʔiɬ ‘distal’, and a

common noun used as a proper name, ʔadad ‘Magpie’:

(490) gʷəl ləʔux%ʷ tiʔiɬ ʔi ʔadad

gʷəl lə=ʔux%ʷ tiʔiɬ ʔi ʔadad

SCONJ PROG=go DIST and magpie

‘and that guy and Magpie went’

(Hess 2006: 37, line 380)

380

An interesting property of coordinate structures formed with anaphoric elements is that when the

referent of the anaphor is plural, the specific identity of one or all of its antecedent may be

repeated as part of the subsequent list of coordinands:

(491) a. ʔəsɬaɬlil tiʔiɬ ʔi sgʷəlub ʔi tiʔiɬ x%əɬx%əɬtəds, tiʔəʔ ʔiɬkʷəlq

ʔəs–ɬaɬlil tiʔiɬ ʔi sgʷəlub ʔi tiʔiɬ x%əɬ–x%əɬtəd–s tiʔəʔ ʔiɬ–kʷəlq

STAT–live DIST and pheasant and DIST DSTR–in.law–3PO PROX PRTV–others

‘they, Pheasant and his brothers-in-law (and) others, lived (there)’

(Hess 2006: 42, line 1)

b. gʷax%ʷəxʷ tiʔiʔiɬ ʔi həbuʔ ʔi tiʔiɬ cədiɬ sč’istxʷs

gʷax%ʷ=əxʷ tiʔ–iʔiɬ ʔi həbuʔ ʔi tiʔiɬ cədiɬ sč’istxʷ–s

walk=now PL–DIST and pigeon and DIST he husband–3PO

‘they, Pigeon and her husband, walked’

(Hess 2006: 35, line 339)

While the most natural translation of these particular examples into English is using an

appositive phrase, the structure in Lushootseed takes the form of a coordinate NP whose first

member is the anaphor and whose subsequent members are introduced by the conjunction ʔi.

Given the optionality of nominal number in Lushootseed, third-person anaphoric heads of such

constructions can be morphologically unmarked (491a) or have overt plural number (491b).

The conjunction ʔi can also form coordinate structures with other types of pronominal

elements such as subject clitics:

(492) a. ɬuɬaɬlil čəɬ ʔi tsiʔiɬ sɬaɬədayʔ ʔal tudiʔ

ɬu=ɬaɬlil čəɬ ʔi tsiʔiɬ sɬa–ɬədayʔ ʔal tudiʔ

IRR=live 1PL.SUB and DIST:FEM ATTN–woman at DIST.DMA

‘that girl and I will live over there’

(Hess & Hilbert 1976: I, 5)

b. tušudxʷ čəɬ ʔi mali

tu=šuɬ–dxʷ–Ø čəɬ ʔi mali

PAST=see–DC–3OBJ 1PL.SUB and Mary

‘Mary and I saw it’

(Bates, Hess & Hilbert: 14)

381

c. tušudxʷ čəɬ əlgʷəʔ ʔi mali

tu=šuɬ–dxʷ–Ø čəɬ əlgʷəʔ ʔi mali

PAST=see–DC–3OBJ 1PL.SUB PL and Mary

‘Mary and I saw them’

(Hess & Hilbert 1976: II, 141, ex. 3)

In these constructions, unlike those in (491), the use of the plural subject clitic (being non-third

person) is obligatory. As can be seen in the example in (492c), the subject-clitic retains its

mobility with respect to linear-ordering and maintains its preferred position as the second

element in the clause, independently of the remainder of the coordinate NP. Presumably, such

expressions in Lushootseed are slightly ambiguous between the first-person dual (‘Mary and I’)

and first-person plural (‘Mary and we’) readings.

The same pattern, referred to as “attraction” by Hess (2006: 37, fn 95; see also Hess &

Hilbert 1976: II, 141), is also seen with the Ø third-person subject clitic, as in (493):

(493) a. cick’ʷəxʷ sʔušəbabdxʷ əlgʷəʔ ʔi tsiʔiɬ dčəgʷas

cick’ʷ=əxʷ sʔušəbabdxʷ Ø əlgʷəʔ ʔi tsiʔiɬ d–čəgʷas

very=now poor.guy 3SUB PL and DIST:FEM 1SG.PO–wife

‘they (including) my wife are very poor’

(Hess 1998: 80, line 72)

b. bədaʔəbəxʷ əlgʷəʔ ʔi tiʔəʔ təkʷtəkʷəlus ʔə tiʔəʔ miʔman’ č’ač’as

bədaʔ–əb=əxʷ Ø əlgʷəʔ ʔi tiʔəʔ təkʷtəkʷəlus ʔə tiʔəʔ

offspring–MD=now 3SUB PL and PROX owl PR PROX

miʔ–man’ č’ač’as

ATTN–small child

‘they had a little boy, [she] and Owl’

(Hess 2006: 3, line 10)

c. tuhuhuyucutəxʷ əlgʷəʔ ʔi tiʔəʔ diʔəʔ suq’ʷaʔs

tu=hu–huyu–t–sut=əxʷ Ø əlgʷəʔ ʔi tiʔəʔ diʔəʔ

PAST=DSTR–made–ICS–REFL=now 3SUB PL and PROX here

suq’ʷaʔ–s

younger.sibling–3PO

‘he prepared himself and his younger brother’

[HM Star Child, line 157]

Just as with the non-third person subject clitics, the overt marking of plurality (using the plural

marker əlgʷəʔ) appears to be obligatory in this type of coordinate construction, although it is not

382

normally an absolute requirement for the marking of plural subjects (or other types of plural

arguments — Section *.*).

Coordination of this type is also possible with imperative subjects:

(494) ʔux%ʷ ʔi tsi sx%aʔhus

ʔux%ʷ Ø ʔi tsi sx%aʔhus

go 2SG.IMP and SPEC:FEM sawbill

‘you and Sawbill go!’

(Hess 2006: 35, line 334)

Unlike the previous examples, however, in the case of imperatives it appears that plural number

is not obligatory for the head of the coordination (which would otherwise be ɬi ‘2PL

IMPERATIVE’). As this is the only example of a coordinate imperative subject in the corpus, it is

not known whether the use of the singular subject is required or is optional (reflecting, say, the

distinction between the English you and Sawbill go! and you go with Sawbill!).

Grammatical attraction of number applies not only to coordinate subjects and objects, but

also is found in the expression of coordinate possessors:

(495) stab kʷi gʷəsgʷaʔs əlgʷəʔ ʔi k’aʔk’aʔ ʔi qaw’qs

stab kʷi gʷə=sgʷaʔ–s əlgʷəʔ ʔi k’aʔk’aʔ ʔi qaw’qs

what REM SBJ=ones.own–3PO PL and crow and raven

‘what belongs to Crow and Raven?’

(Hess 1998: 75, line 259)

Note that, as in many of the preceding examples, the list of possessors in the string of

coordinands is exhaustive.

A similar construction to that seen in (495) is also seen with s=nominals (Section 7.4.2.1)

whose possessors are coordinate, as in (496):

383

(496) xʷiʔəxʷ kʷi ɬadsux%ʷubil ʔi tiʔəʔ biʔbədaʔ

xʷiʔ=əxʷ kʷi ɬu=ad=s=ʔu–x%ʷubil ʔi tiʔəʔ biʔ–bədaʔ

NEG=now REM IRR=2SG.PO=NM=PFV–be.quiet and PROX ATTN–offspring

‘you and your little baby will never be quiet’

141

(Hess 2006: 8, line 140)

The parallel between (495) and (496) is unsurprising, given that in most other respects the

possessors of s=nominals follow the same pattern as possessors of ordinary NPs.

7.4 Relative, headless relative, and nominalized clauses

Target

Clause type

Subject

relative clause

Direct object

3 inanimate object

3 animate object

relative clause

s=nominal

Oblique object

s=nominal

Adjunct

dəxʷ=nominal

Table 78: Types of modifying clauses

7.4.1 Modifying and headless relative clauses

Relative clauses in Lushootseed take the form of a fully inflected clause without any overt

marker of subordination. When used as a modifier, the relative most usually follows its head:

(497) a. tsiʔəʔ bədaʔ ʔə tiʔəʔ tusbiaw ʔuʔatəbəd

tsiʔəʔ bədaʔ ʔə tiʔəʔ tu=sbiaw ʔu–ʔatəbəd

PROX:FEM offspring PR PROX PAST=coyote PFV–be.dead

‘the daughter of Coyote who has died’

(Hess 1998: 97, line 181)

141

This is the gloss given in Hess (2006: 239); a more literal translation would be ‘you and the little child will never

be quiet’. It is not clear if the use of the second-person possessive your actually reflects some feature of Lushootseed

syntax that requires that the child be ‘your’ child in this type of construction or simply reflects the contextual

knowledge of the translator that the child in question is indeed the addressee’s.

384

b. huy, ʔibəšəxʷ tiʔəʔ sʔušəbabdxʷ ʔuɬiltəb

huy ʔibəš=əxʷ tiʔəʔ sʔušəbabdxʷ ʔu–ɬil–t–əb

INTJ travel=now PROX humble.fellow PFV–given.food–ICS–PASS

‘then this humble fellow to whom food had been given traveled’

(Hess 1998: 82, line 112)

c. ʔušudxʷ čəxʷ ʔu tiʔiɬ stiqiw ʔuč’ax%ʷč’axʷatəb ʔə ti x%ik’ʷ ləgʷəb

ʔu–šuɬ–dxʷ čəxʷ ʔu tiʔiɬ stiqiw ʔu–č’ax%ʷ–č’axʷa–t–əb

PFV–see–DC 2SG.SUB INT DIST horse PFV–DSTR–club–ICS–PASS

ʔə ti x%ik’ʷ ləgʷəb

PR SPEC ugly youth

‘did you see the horse that was beaten by the ugly youth?’

(Hess & Hilbert 1976: II, 127)

In (497a) we see the intransitive verb ʔatəbəd ‘be dead’ following and modifying the noun

sbiaw ‘coyote’. (497b) shows the passive form of the verb ɬild ‘give food’ modifying the

preceding noun, sʔušəbabdxʷ ‘humble fellow’, while (497c) gives a more complex example of a

passivized verb, č’ax&ʷč’axʷatəb ‘be beaten’, in a relative clause containing an overt agentive

complement, ʔə ti x&ik’ʷ ləgʷəb ‘by the ugly youth’.

Post-nominal modifying relative clauses are often introduced by a determiner:

(498) huˑy, dᶻux%ʷatəxʷ ʔə tiʔəʔ x%ʷul’ul’əxʷ p’q’ac tiʔəʔ tusuʔələds

huy dᶻux%ʷat=əxʷ ʔə tiʔəʔ x%ʷul’–ul’=əxʷ p’q’ac

SCONJ vomit=now PR PROX EXC–only=now rotten.wood

tiʔəʔ tu=s=ʔu–ʔələd=s

PROX PAST=NM=PFV–eat=3PO

‘then he vomited up the rotten wood that he had eaten’

(Hess 1998: 88, line 277)

In this particular example, the determiner tiʔəʔ preceding the relative clause tusuʔələds, was not

part of the sentence as it was given by the story-teller, but was added later as a “correction”

during the transcription of this particular story. However, both the sentence with and without the

determiner are judged as grammatical by consultants. The use versus non-use of the determiner

seems to correspond to differences in register or formality.

385

Although post-nominal modifying relative clauses are more frequent, relative clauses based

on monovalent intransitive verbs or on verbs without overt objects can precede the nouns they

modify:

(499) a. tucutəb čəɬ ʔə tiʔiɬ tuʔatəbəd sbiaw

tu=cut–t–əb čəɬ ʔə tiʔiɬ tu=ʔatəbəd sbiaw

PAST=say–ICS–PASS 1PL.SUB PR DIST PAST=be.dead coyote

‘we were told by Coyote who died’

(Hess 1998: 98, 193)

b. ƛ’ulək’ʷəd tiʔəʔ ƛ’uɬčiltub haʔɬ sʔəɬəd

ƛ’u=lək’ʷ–əd tiʔəʔ ƛ’u=ɬčil–txʷ–b haʔɬ sʔəɬəd

HAB=eat–ICS PROX HAB=arrive–ECS–PASS good food

‘she would eat the good food that would be brought’

(Bates, Hess & Hilbert 1994: 105)

The first example, (499a), shows the intransitive verb ʔatəbəd ‘be dead’ (cf. (497a) above)

modifying the following noun, sbiaw ‘coyote’, while the second in (499b) shows the passive

form of a transitive verb modifying haʔɬ sʔəɬəd ‘good food’ (itself composed of a bare modifier

and a noun). Note that the structure in (499b), resembles an internally-headed relative clause.

In addition to their uses as nominal modifiers, relative clauses are also very frequently

(perhaps more frequently) used as full referential expressions without an overt nominal head.

These headless relatives take the form of a full clause introduced by a determiner, and are

interpreted as referring to one of the arguments of the subordinated verb phrase:

(500) a. ƛ’əƛ’iq’šəd tə ƛ’uʔux%ʷ

ƛ’əƛ’iq’šəd tə ƛ’u–ʔux%ʷ

sapsucker NSPEC HAB–go

‘the one who would go was Sapsucker’

(Hess 2006: 18, line 164)

b. skəyu təɬ tiʔiɬ ʔucucuuc čələp

skəyu təɬ tiʔiɬ ʔu–cut–cut–c čələp

corpse truly DIST PFV–DSTR–say–ALTV 2PL.SUB

‘that which you guys are talking about is truly a ghost’

(Hess 1998: 94, line 107)

386

The first two examples here show a particularly frequent use of the headless relative clause as the

subject of a nominal predicate. These structures are identical to the relative clause constructions

illustrated in (497) and (499) except for the introductory determiner (cf. 500a and ti ƛ’əƛ’iq’šəd

ƛ’uʔux&ʷ ‘the sapsucker who would go’). The subject phrase of (500a) is interpreted as referring

to the syntactic subject of the embedded verb (‘sapsucker’), while the subject of (500b) refers to

the direct object of the embedded verb cuuc ‘speak about something’.

As the examples in (500) show, relativization is possible for subjects and direct objects,

although there are strict limitations on which argument of the verb can be relativized in a

particular syntactic or discourse context. All other things being equal, in clauses with a third-

person subject and a third-person object, only the subject can be relativized:

(501) a. ʔušudxʷ čəɬ ti č’ač’as ʔutəsəd tiʔiɬ stubš

ʔu–šuɬ–dxʷ čəɬ ti č’ač’as ʔu–təs–əd tiʔiɬ stubš

PFV–see–DC 1PL.SUB SPEC child PFV–hit–ICS DIST man

‘we saw the boy that hit the man’

*‘we saw the boy that the man hit’

(Hess & Hilbert 1976: II, 125)

b. ʔušudxʷ čəd ti sqʷəbayʔ ʔuč’axʷatəb ʔe tiʔiɬ č’ač’as

ʔu–šuɬ–dxʷ čəd ti sqʷəbayʔ ʔu–č’axʷa–t–əb ʔe tiʔiɬ č’ač’as

PFV–see–DC 1SG.SUB SPEC dog PFV–clubbed–ICS–PASS PR DIST child

‘I see the dog that was hit with a club by the boy’

(Hess & Hilbert 1976: II, 124)

(501a) gives an example of an object-centred modifying relative clause with a third-person

subject and a third-person object; in this case, the only interpretation of the sentence possible is

that of a subject-centred relative clause. When the object-centred reading is desired, it is

necessary to passivize the embedded clause as in (501b). The same holds for headless relative

clauses like those shown in (502):

387

(502) a. wiw’su tiʔəʔ ʔučalad tiʔəʔ sqʷəbayʔ

wiw’su tiʔəʔ ʔu–čala–d tiʔəʔ sqʷəbayʔ

children PROX PFV–chased–ICS PROX dog

‘the ones who chased the dog [are] the children’

*‘the one who the dog chased [are] the children’

b. sqʷəbayʔ ti ʔučalatəb ʔə tiʔiɬ wiw’su

sqʷəbayʔ ti ʔu–čala–t–əb ʔə tiʔiɬ wiw’su

dog SPEC PFV–chased–ICS–PASS PR DIST children

‘the one who was chased by the children is the dog’

(Hess 1995: 99)

In (502a), the only interpretation open to the sentence is the one where the headless relative

clause identifies the subject of the embedded verb, in spite of the fact that the opposite

interpretation, where the dog chases the children, is semantically and pragmatically quite

plausible. Again, where this is the desired interpretation, the embedded clause appears in the

passive, as in (502b).

Nevertheless, when discourse context leaves no room for ambiguity as to the syntactic roles

of the third-person arguments of the verb in the embedded clause, object-centred relative clauses

are possible:

(503) a. taxʷčəɬəb sʔəɬəd ʔə tiʔəʔ diʔəʔ stawixʷʔɬ tasčəbaʔəd tul’ʔal tudiʔ čaʔkʷ

tu=ʔas–dxʷ–čəɬ–əb sʔəɬəd ʔə tiʔəʔ diʔəʔ stawixʷʔɬ

PAST=STAT–CTD–make–DSD food PR PROX here children

tu=ʔas–čəbaʔ–əd tul’–ʔal tudiʔ čaʔkʷ

PAST=STAT–laden–ICS CNTRFG–at DIST.DMA waterward

‘[Basket Ogress] wanted to make food of the children she carried up from the water’

[DM Basket Ogress, line 73]

b. tuɬiltubuɬ ʔə ti sqigʷəc tuq’ʷəx%ʷəd

tu=ɬil–txʷ–buɬ ʔə ti sqigʷəc tu=q’ʷəx%ʷ–əd

PAST=give.food–ECS–1PL.OBJ PR SPEC deer PAST=butchered–ICS

‘he gave us the deer which he had butchered’

(Hukari 1977: 53)

Even in such cases, object-centred relatives like these are unusual, the more common pattern

being for the embedded verb being used in the passive voice, promoting the object to subject and

allowing for a subject-centred relative clause.

388

When the subject of the embedded clause is first- or second-person, only object-centred

relative clauses are possible:

(504) a. ʔušudxʷ čəɬ ti č’ač’as ʔutəsəd čəd

ʔu–šuɬ–dxʷ čəɬ ti č’ač’as ʔu–təs–əd čəd

PFV–see–DC 1PL.SUB SPEC child PFV–hit–ICS 1SG.SUB

‘we saw the boy that I hit’

(Hess & Hilbert 1976: II, 125)

tiʔəʔ dəč’uʔ syəyəhub, q’iq’x%ʷuʔ tiʔəʔ ʔuʔilid čəd ʔal tiʔəʔ səɬax%il

tiʔəʔ dəč’uʔ syəyəhub q’iq’x%ʷuʔ tiʔəʔ ʔu–ʔili–d čəd

PROX one story short PROX PFV–sing–ECS 1SG.SUB

ʔal tiʔəʔ səɬax%il

PR PROX night

‘this one story, that which I have sung this evening, is short’

[HM Star Child, line 194]

This is almost certainly a syntactic restriction, as the subject-markers are not themselves

nominals and so cannot head an NP or be modified by a relative clause. There are no examples in

the present corpus of first- or second-person pronouns heading a relative clause construction, but

there are numerous examples of pronouns functioning as predicates of sentences with headless

relative clause subjects. In these cases, the pronoun and the headless relative express the same

semantic actant, and the verb embedded in the headless relative clause is in the third-person:

(505) a. ʔəca tiʔəʔ ləčalad tə sqʷəbayʔ

ʔəca tiʔəʔ lə=čala–d tə sqʷəbayʔ

I PROX PROG=chased–ICS NSPEC dog

‘the one who is chasing the dog [is] me’

b. dibəɬ ti ʔut’uc’utəb ʔə tiʔiɬ šəbad

dibəɬ ti ʔu–t’uc’u–t–əb ʔə tiʔiɬ šəbad

we SPEC PFV–shot–ICS–PASS PR DIST enemy

‘the ones who were shot by the enemy [are] us’

(Hess 1995: 99)

c. gʷəl dəgʷi kʷi ɬukʷədatəb dᶻixʷ

gʷəl dəgʷi kʷi ɬu=kʷəda–t–əb dᶻixʷ

then you REM IRR=held–ICS–PASS first

‘the one who will be taken first [is] you’

[LA Basket Ogress, line 26]

389

In these constructions, the headless relative clauses are obligatorily subject-centred, or at least

are uniformly so in the present corpus. When expression of the PATIENT or ENDPOINT of the event

is the sentence predicate, the subject-phrase appears in the passive voice, as in (505c).

7.4.2 Nominalized clauses

Although Lushootseed does not allow the relativization of oblique objects or adjuncts, it

creates the structural equivalent of oblique- and adjunct-centred relative clauses through the

formation of gerund- or participle-like constructions using one of the two nominalizing proclitics

— s= or dəxʷ=. Such clausal nominals have essentially the same internal syntax as matrix

clauses in terms of the valency and transitivity of the nominalized predicate; however, an

important difference between the two clause types is that nominalized clauses mark their subjects

with the possessive series of subject-marking clitics (Section 8.1.1), as shown in (506):

(506) a. x%ʷul’ čəd ɬuləʔux%ʷtxʷ tiʔəʔ ɬadsʔəɬtxʷ

x%ʷul’ čəd ɬu=lə=ʔux%ʷ–txʷ tiʔəʔ ɬu=ad=s=ʔəɬ–txʷ

only 1SG.SUB IRR=PROG=go–ECS PROX IRR=2SG.PO=NM=eat–ECS

‘I will just be taking [them] what you will feed [them] with’

(Hess 1998: 58, line 56)

b. huyəxʷ tiʔiɬ dsyəhubtubicid, sʔiab dsyaʔyaʔ

huy=əxʷ tiʔiɬ d=s=yəhub–txʷ–bicid sʔiab d–syaʔyaʔ

be.done=now DIST 1SG.PO=NM=recite–ECS–2SG.OBJ noble 1SG.PO–friend

‘my telling to you is finished now, my noble friend’

(Hess 1995: 142, line 51)

In (506a) the subject of the nominalized clause sʔəɬtxʷ ‘feeding someone’ (based on the

transitive verb ʔəɬtxʷ ‘feed someone with something’) is expressed by the second-person singular

possessive subject clitic, ad=. Similarly, in (506b) the subject of syəhubtubicid ‘telling to you’

is expressed by the first-person singular subject proclitic, d=. dəxʷ=nominals with first- and

second-person subjects also use elements from the possessive-subject paradigm:

390

(507) a. ləliʔəxʷ tiʔəʔ cəxʷuʔibəš

ləliʔ=əxʷ tiʔəʔ d=dəxʷ=ʔu–ʔibəš

different=now PROX 1SG.PO=ADNM=PFV–travel

‘where I am traveling is different now’

(Hess 2006: 27, line 128)

b. ƛ’uləbəlx%ʷ ʔal tiʔiɬ čad dəxʷʔaləp

ƛ’u=lə=bəlx%ʷ ʔal tiʔiɬ čad dəxʷ=ʔa=ləp

HAB=PRG–go.by PR DIST where ADNM=be.there=2PL.PO

‘he goes by there where you guys come from’

(Hess 2006: 66, line 592)

(507a) shows the first-person singular proclitic d= marking the subject of the nominalized clause

cəxʷuʔibəš ‘where I travel’. The next example in (507b) contains a nominalized clause with a

second-person plural subject acting as relative clause modifying čad ‘where’, which is in turn

contained within an prepositional phrase acting as a locative adverbial modifier.

note on order of nominal wrt head, give normal case, preceding, then

If a pre-posed relative clause is complex — that is, it contains elements other than the verb —

the additional elements can appear following the modified noun (although this is extremely rare):

(508) x%ʷul’əxʷ ʔukʷəd(d)xʷ tiʔəʔ diʔəʔ bəčəɬs bədaʔ tul’ʔal tiʔəʔ tuscəcikʷ ʔup’ic’id

x%ʷul’=əxʷ ʔu–kʷəd–dxʷ tiʔəʔ diʔəʔ bə=s=čəɬ=s bədaʔ

only=now PFV–take–DC PROX here ADD=NM=make=3PO child

tul’–ʔal tiʔəʔ tu=scəcikʷ ʔu–p’ic’i–d

CNTRFG–at PROX PAST=DIAPER PFV–wrung–ICS

‘she just took this child made from the wrung out diaper’

[HM Star Child 69, line 69]

Here, the noun bədaʔ ‘child’ is modified by the nominalized clause čəɬs bədaʔ tul’ʔal tiʔəʔ

tuscəcikʷ ʔup’ic’id ‘made from a wrung-out diaper’, within which it corresponds to a nominal

predicate complement (Section 8.2.5). The noun itself appears following the verb of the

embedded clause but preceding the adverbial adjunct phrase tul’ʔal tiʔəʔ tuscəcikʷ ʔup’ic’id

‘from a wrung-out diaper’. This is an unusual construction (more common would be to post-pose

the entire nominalized clause) and strongly resembles an internally-headed relative clause (see

also example 499b above).

391

When the subject of either type of nominalized clause is third-person, it shows the same

range of variation as the expression of the third-person possessor does, using the possessive

subject enclitic =s if there is no overt subject NP, or otherwise making use of a periphrastic

possessive construction:

(509) a. x%ʷul’ p’aƛ’aƛ’ tiʔiɬ sʔabyids tiʔiɬ č’ƛ’aʔ

x%ʷul’ p’aƛ’aƛ’ tiʔiɬ s=ʔab–yi–d=s tiʔiɬ č’ƛ’aʔ

only worthless DIST NM=extend–DAT–ICS=3PO DIST rock

‘what he gives to that rock is simply worthless’

(Hess 1995: 148, line 32)

b. tiʔiɬ tusʔukʷukʷ ʔə tə wiw’su

tiʔiɬ tu=s=ʔukʷukʷ ʔə tə wiw’su

DIST PAST=NM=play PR NSPEC children

‘what the children were playing with’

(Hess 1998: 89, line 299)

In (509a) the subject is realized as simply the third-person possessive subject marker, =s, while

in (509b) the subject is an overt NP, tə wiw’su ‘the children’, and so the periphrastic possessive

construction with ʔə is used. The same two patterns are also observed with dəxʷ=nominals:

(510) a. ƛ’al’ bədiɬ dəxʷʔa ʔə tiʔiɬ dəxʷʔəy’dubs ʔə tiʔiɬ sgʷəlub

ƛ’al’ bə=diɬ dəxʷ=ʔa ʔə tiʔiɬ dəxʷ=ʔəy’–dxʷ–b=s

also ADD=FOC ADNM=be.there PR DIST ADNM=find–DC–PASS=3PO

ʔə tiʔiɬ sgʷəlub

PR DIST pheasant

‘it was the very same place where they had been found by Pheasant’

(Hess 1998: 85, line 187)

b. ʔəsɬaɬlil tiʔiɬ ʔaciɬtalbixʷ dəxʷʔa ʔə tiʔacəc sbiaw

ʔəs–ɬaɬlil tiʔiɬ ʔaciɬtalbixʷ dəxʷ=ʔa ʔə tiʔacəc sbiaw

STAT–live DIST person ADNM=be.there PR UNQ coyote

‘people were living where Coyote was’

(Hess 1998: 91, line 1)

Again, here we see the use of the subject enclitic =s when there is no overt subject NP present

(510a), and the periphrastic construction with ʔə used with an overt NP (510b). An alternative

construction, the use of =s with a full NP, is also possible, although it is largely confined to

nominalized passive constructions. This will be discussed in more detail in Section 8.1.1 below.

392

Although the nominalizing proclitics s= and dəxʷ= appear adjacent to the stem, they are —

like many other elements in Lushootseed grammar — in fact phrase-level proclitics rather than

word-level prefixes. As such, they are obligatorily attached to the first element in the

nominalized clause, whether or not this element is the sentence predicate, as shown in (511):

(511) a. ʔa əw’ə sixʷ tiʔiɬ adsuhuy ti ƛ’ubəstiləbsəxʷ ƛ’ubəšəq

ʔa əw’ə sixʷ tiʔiɬ ad=s=ʔu–huy

be.there PTCL PTCL DIST 2SG.PO=NM–PFV–be.done

ti ƛ’u=bə=s=tiləb=s=əxʷ ƛ’u=bə=šəq

SPEC HAB=ADD=NM–suddenly=3PO=now HAB=ADD=high

‘There is something you do to make it suddenly go high again’

(lit. ‘what you do [so that] it suddenly goes high again is there [i.e., exists]’)

(Hess 2006: 26, line 102)

b. ʔəsʔəx%id əw’ə tsiʔəʔ adčəgʷas dəx%ʷul’s ʔubak’ʷacut tiʔəʔ qədxʷs

ʔəs–ʔəx%id əw’ə tsiʔəʔ ad–čəgʷas dəxʷ=x%ʷul’=s

STAT–what.happen PTCL PROX:FEM 2SG.PO–wife ADNM=only=3PO

ʔu–bak’ʷa–t–sut tiʔəʔ qədxʷ–s

PFV–move.quickly–ICS–REFL PROX mouth–3PO

‘what is the matter with your wife that her mouth is just a-goin’?’

(Hess 2006: 4, line 22)

The sentence in (511a) shows a nominalized clause ti ƛ’ubəstiləbsəxʷ ƛ’ubəšəq ‘its suddenly

going high again’ acting as the subject of the verb huy ‘be done, be made, be finished’ (itself

nominalized). Because the initial element of the clause is an adverb, tiləb ‘suddenly’, the

nominalizing proclitic s= is attached to this word instead of to the clausal predicate, šq ‘high’, as

are the other clitics ƛ’u= ‘habitual’, bə= ‘additive’, =s ‘third-person possessive subject’, and

=əxʷ ‘now’. Similarly, in (511b) we see the proclitic dəxʷ= attached to an adverb, x&ʷul’ ‘only’,

which is also the initial element in the embedded clause (cf. the matrix clause, x&ʷul’ ʔubak’ʷacut

tiʔəʔ qədxʷs ‘s/he would only move her/his mouth’). Note also that in both cases the subject

clitic goes along with the nominalizing proclitic and appears on the adverb rather than on the

verb. In a few cases, the nominalizer appears on both the adverb and the clausal predicate:

393

(512) … tul’ʔal kʷi sdᶻixʷs tushuyutid ʔə sp’ic’ikʷ ʔi sɬukʷalb

tul’ʔal kʷi s=dᶻixʷ=s tu=s=huyu–t–id ʔə sp’ic’ikʷ

PR REM NM=first=3PO PAST=NM=be.done–ICS–PASS.SBRD PR Diaper.Child

ʔi sɬukʷalb

and moon

‘… because of what was first done by Diaper Child and Moon’

[DS Star Child, line 371]

In this respect, the nominalizing clitics behave like other phrase-level proclitics in that they can

be iterated on more than one element of the phrase.

Although the internal syntax of both types of nominalized clause is similar, their distributions

are quite different. The nominalizer s= is used primarily to form the equivalent of oblique-

centred relative and headless relative clauses. The proclitic dəxʷ=, on the other hand, has not

only this function, but is also used to form a variety of adverbial expressions. The uses of the two

types of clausal nominals will be discussed further in the remainder of this section.

7.4.2.1 s=nominals

The more common of the two nominalizers, s=, has two principal functions. The first of

these is to form the equivalent of oblique-centred relative clauses. As with relative clauses, these

may or may not have an overt nominal head:

(513) a. gʷəl ləbəčad tiʔəʔ cədiɬ sxʷiʔxʷiʔs səsɬiltəbs

gʷəl lə=bəčad tiʔəʔ cədiɬ sxʷiʔxʷiʔ–s s=ʔəs–ɬil–t–əb=s

then PROG=fallen–ICS PROX it game–3PO NM=STAT–give.food–ICS–PASS=3PO

‘then he set down his game that he had been given as food’

(Hess 1998: 82, line 124)

142

b. day’əxʷ haʔɬ tiʔəʔ sʔəsq’ʷəlb ʔə tiʔacəc bəsčəb

day’=əxʷ haʔɬ tiʔəʔ s=ʔəs–q’ʷəl–b ʔə tiʔacəc bəsčəb

really=now good PROX NM=STAT–ready.to-eat–CSMD PR UNQ mink

‘that roasting [salmon] of Mink’s is really good’

(Hess 1998: 66, line 42)

142

This line appears in Hess (1998) with an editorial addition of the determiner tiʔəʔ introducing the relative clause,

although this is not on the spoken version on tape. Both sentence as it is spoken by the narrator and as it appears in

the written text are grammatical.

394

(513a) shows the nominalization of the passive form of a transitive verb, ɬild ‘give food’. This

nominalized clause modifies the noun sxʷiʔxʷiʔ ‘game’, which corresponds to an oblique object

of the embedded clause (cf. ʔəsɬiltəb ʔə ti sxʷiʔxʷiʔ ‘he is given game’). In (513b), the reference

of the nominal is the oblique object of the intransitive verb q’ʷəlb ‘roast oneself something’ (cf.

q’ʷəlb ʔə ti sʔuladxʷ ‘he roasted himself a salmon’). It should be noted that while nouns

corresponding to oblique objects can be modified by s=nominalizations, nouns corresponding to

agentive complements of passives can not: in these cases, the active form of the verb is used to

form a subject-centred relative clause.

Nominalized clauses fill the same syntactic roles as relative clauses, either modifying a noun

(513a) or acting as a syntactic argument (513b). Like a relative clause, a modifying nominalized

clause can appear either following its nominal head as in (513a) or preceding it as in (514):

(514) tiləb ʔugʷəx%agʷil tiʔəʔ tusčəbaʔs kʷagʷičəd

tiləb ʔu–gʷəx%–agʷil tiʔəʔ tu=s=čəbaʔ=s kʷagʷičəd

suddenly PFV–get.loose–AUTO PROX PAST=NM=laden=3PO elk

‘suddenly the elk that he’d carried on his back got loose’

(Hess 1998: 87, line 253)

In this example, the nominalized clause tusčəbaʔs ‘what he’d been carrying on his back’

precedes its modificand, kʷagʷičəd ‘elk’, which corresponds to the oblique object of the bivalent

intransitive verb čəbaʔ ‘be loaded down with something’.

The second function of the s=nominalizer is to form sentential nominals (Beck 2000a),

nominalizations of entire clauses whose reference is the event rather than a particular event-

participant. Compare the nominalized clauses in (509) with those in (515):

(515) a. tul’t’aq’t tiʔəʔ suʔəƛ’ ʔə tiʔəʔ qʷuʔ

tul’t’aq’t tiʔəʔ s=ʔu–ʔəƛ’ ʔə tiʔəʔ qʷuʔ

waterward PROX NM=PFV–come PR PROX water

‘the coming of the water is waterward’

(Hess 1998: 69, line 108)

395

b. ʔəsluud əlgʷəʔ tiʔiɬ suƛ’əladiʔs ʔal kʷədiʔ t’aq’t

ʔəs–luh–d əlgʷəʔ tiʔiɬ s=ʔu–ƛ’əladiʔ=s ʔal kʷədiʔ t’aq’t

STAT–heard–ICS PL DIST NM=PFV–make.noise=3PO at REM:DMA inland

‘they heard her making noise over there on shore’

(Hess 2006: 17, line 134)

The nominalized clauses in these examples refer to entire events — the coming of the water in

(515a) and the making of a noise in (515b). In neither case is the reference of the nominalized

clause an argument of the verb (cf. (509), where the nominalized clause refers to an oblique

argument of the verb). The syntactic role taken by such sentential nominals is restricted largely

to that of syntactic argument, as in the examples in (515), or as the object of a preposition in an

adverbial phrase (Section 8.2.7):

(516) ʔudᶻix%ič ʔal tiʔəʔ tusušəɬs hud

ʔu–dᶻix%•ič ʔal tiʔəʔ tu=s=ʔu–šəɬ=s hud

PFV–break.off•cover at PROX PAST=NM=PFV–make=3PO wood

‘he broke off the covering [i.e., cedar bark] when he was making firewood’

(Hess 1998: 72, line 172)

As in (515), the reference of the nominalized clause here is the event, making firewood, rather

than a particular event-participant. The contrast between an argument-centred headless relative

clause, an argument-centred s=nominal, and a headless sentential nominal using s= is illustrated

by the following pair of sentences:

(517) a. ʔax%adus əw’ə tiʔəʔ ʔəsɬaɬlil ʔal tiʔəʔ diʔəʔ sbabdil

ʔax%adus əw’ə tiʔəʔ ʔəs–ɬaɬlil ʔal tiʔəʔ diʔəʔ sbabdil

basket.ogress PTCL PROX STAT–live PR PROX here little.mountain

‘the one who lived in those little mountains was Basket Ogress’

[AW Basket Ogress, line 44]

b. sp’ic’ikʷ tə sdaʔs

sp’ic’ikʷ tə s=daʔ=s

Diaper.Child NSPEC NM=named=3PO

‘what he is named [is] Diaper Child’

[MW Star Child, line 118]

David Beck 10-2-7 2:19 PM

Comment: I think this is mis-analyzed. It is

probably np-named-3po

396

c. xʷiʔ ləhaʔkʷ tiʔəʔ səsɬaɬlils

xʷiʔ lə=haʔkʷ tiʔəʔ s=ʔəs–ɬaɬlil=s

NEG NEGP=be.long.time PROX NM=STAT–live=3PO

‘their living [together] has not for a long time’

(Hess 2006: 3, line 6)

In (517a), the referent of the headless relative clause, tiʔəʔ ʔəsɬaɬlil ʔal tiʔəʔ diʔəʔ sbabdil ‘the

one who lived in those little mountains’ is its syntactic subject, ʔax&adus ‘Basket Ogress’. The

referent of the s=nominal in (517b), tə sdaʔs, however, is an oblique object of the verb √daʔ ‘be

named something’, while in (517c), the nominalized clause refers to the event səsɬaɬlils ‘their

living there’. In these last two cases, the nominalizing proclitic s= is required. The same

contrast is observed with transitive verbs such as čalad ‘chase something’:

(518) a. wiw’su tiʔəʔ ʔučalad tiʔəʔ sqʷəbayʔ

wiw’su tiʔəʔ ʔu–čala–d tiʔəʔ sqʷəbayʔ

children PROX PFV–chased–ICS PROX dog

the ones the dog chased [are] the children’

b. sqʷəbayʔ ti ʔučalatəb ʔə tiʔiɬ wiw’su

sqʷəbayʔ ti ʔu–čala–t–əb ʔə tiʔiɬ wiw’su

dog SPEC PFV–chased–ICS–PASS PR DIST children

the one who was chased by the children [is] the dog’

(Hess 1995: 99)

c. ʔəsx%icil tiʔiɬ č’ƛ’aʔ ʔə tiʔiɬ sučalads tiʔiɬ sbiaw

ʔəs–x%ic–il tiʔiɬ č’ƛ’aʔ ʔə tiʔiɬ s=ʔu–čala–d=s tiʔiɬ sbiaw

STAT–angry–INCH DIST stone PR DIST NM=PFV–chased–ICS=3PO DIST coyote

‘Stone was angry as he chased Coyote’

(Hess 1995: 148, line 30)

Again, the desired reference for the headless relative clauses in (518a) and (b) are the subjects

of the embedded clause, whereas the nominalized clause in (518c) refers to the whole event.

Although sentential nominals do not refer to event-participants, they are nonetheless

nominals and fill the same nominal syntactic roles in sentences that headless relative clauses and

oblique-centred s=nominals do — that is, as arguments of verbs and objects of prepositions.

Both types of s=nominals also appear occasionally as sentence-predicates (see Section 8.3.1

below), particularly in conjunction with the focalizing element diɬ (11.2) and with verbs that take

397

subject-complements (8.2.5). Additionally, sentential nominals can act as adverbial clauses (9.5).

The reader is referred to the relevant sections for detailed discussion of these constructions.

7.4.2.2 dəxʷ=nominals

The second nominalizing proclitic, dəxʷ=, is used primarily for the nominalization of adjunct

and circumstantial expressions, particularly (but not exclusively), instruments, locations,

motivations, and manners, and will be referred to here as an adjunctive nominalizer. Like

s=nominals, dəxʷ= creates nominal expressions with the same distribution as relative and

headless relative clauses. The following examples illustrate adjunctive nominals referring to

instruments in the syntactic roles of predicate complement (519a) and relative clause (519b):

(519) a. gʷəl diɬ dəxʷšəɬtəbiɬədtubs

gʷəl diɬ dəxʷ=šəɬ–təbiɬəd–txʷ–b=s

then FOC ADNM=make–rope–ECS–PASS=3PO

‘then these were made into rope’

(Hess 1998: 80, line 81)

b. kʷədad tiʔəʔ qʷuʔ dəxʷuc’agʷačiʔbs ʔə tiʔəʔ sbaɬs

kʷəda–d tiʔəʔ qʷuʔ dəxʷ=ʔu–c’agʷ–ačiʔ–b=s ʔə tiʔəʔ s=baɬ=s

taken–ICS PROX water ADNM=PFV–wash–hand–MD=3PO PR PROX NM=cure=3PO

‘he took some water that he washed his hands with for the curing ceremony’

143

(Hess 1998: 57, line 36)

The examples in (520) show adjunctive nominals expressing locations, used as an argument and

the object of a preposition, respectively:

(520) a. hay gʷəl ɬuhuyutəbəxʷ č’it tiʔiɬ dəxʷʔibəš čəɬ

hay gʷəl ɬu=huyu–t–əb=əxʷ č’it tiʔiɬ dəxʷ=ʔibəš čəɬ

INTJ then IRR=be.done–ICS–PASS=now nearby DIST ADNM=travel 1PL.PO

‘and then where we are traveling [to] will be made near’

(Hess 1995: 147, line 10)

143

This sentence appears in the original source with an editorial amendment of tiʔiɬ ‘distal determiner’ inserted

between qʷuʔ ‘water’ and the nominalized clause. This word is not on the original recording and the sentence is

correct without it. The form given here corresponds more closely to the structure of the English gloss.

398

b. hay gʷəl bəɬčiləxʷ dxʷʔal tiʔiɬ bədəxʷɬčils

hay gʷəl bə=ɬčil=əxʷ dxʷ–ʔal tiʔiɬ bə=dəxʷ=ɬčil=s

INTJ then ADD=arrive=now CNTRPT–at DIST ADD=ADNM=arrive=3PO

‘and then he arrived at where he arrived’

(Hess 1998: 66, line 22)

These last two types of adjunctive nominals are the most noun-like in the sense that their

referents are physical entities (objects and locations). However, dəxʷ= is also used to form a

wide variety of other expressions, one of the most common being expressions of motive:

(521) a. sɬild čəɬ ti dəgʷi tiʔiɬ dəxʷuwiliq’ʷid čəɬ

s=ɬil–d čəɬ ti dəgʷi tiʔiɬ dəxʷ=ʔu–wiliq’ʷi–d čəɬ

NM=give.food–ICS 1PL.PO SPEC you DIST ADNM=question–ICS 1PL.PO

‘the reason that we questioned you [is] that we are giving you this food’

144

(Hess 1998: 80, line 69)

b. ʔa tiʔiɬ adəxʷx%əɬ, x%ənimulicaʔ

ʔa tiʔiɬ ad=dəxʷ=x%əɬ x%ənimulicaʔ

be.there DIST 2SG.PO=ADNM=sick x%ənimulicaʔ

‘there is a reason that you are sick, x%ənimulicaʔ’

(Hess 1998: 57, line 26)

In both of these sentences, the adjunctive nominal serves as the syntactic subject.

Another type of adjunctive nominal expresses manner:

(522) a. gʷəl diɬ tiʔəʔ cəxʷkʷədxʷ tiʔəʔ dsʔəɬəd

gʷəl diɬ tiʔəʔ d=dəxʷ=kʷəd–dxʷ tiʔəʔ d=s=ʔəɬəd

then FOC PROX 1SG.PO=ADNM=take–DC PROX 1SG.PO=NM=eat

‘than how I managed to get my food was that [way]’

(Hess 1998: 83, line 164)

b. hay, diɬ dəxʷhuyuds ʔušəbabdxʷ tsiʔə ʔalš ʔə tiʔəʔ suʔəɬadəps

hay diɬ dəxʷ=huyu–d=s ʔušəbabdxʷ tsiʔə ʔalš ʔə

INTJ FOC ADNM=be.done–ICS=3PO poor.dear PROX:FEM cross.sex.sibling PR

tiʔəʔ s=ʔu–ʔəɬadəp=s

PROX NM=PFV–give.feast=3PO

‘so, that is how he made his sister poor when she gave a feast’

(Hess 1998: 63, line 66)

144

This sentence appears in (Hess 1998) without the initial s=nominalizer on the sentence predicate ɬild ‘give food’.

399

In the first sentence here, the clausal nominal serves as the syntactic subject of the copula diɬ,

while in the second it serves as the complement of the same element (see Sections 8.2.5 and 11.2

below for discussion of these structures).

A less commonly-attested use of adjunctive nominals is to indicate purpose:

(523) a. huˑy, huyudəxʷ tiʔiɬ sx%əɬ dxʷʔal kʷi dəxʷx%əɬ ʔə tsiʔəʔ ʔalšs

huy huyu–d–əxʷ tiʔiɬ s–x%əɬ dxʷ–ʔal kʷi dəxʷ=x%əɬ

INTJ be.done–ICS–now DIST NP–sick CNTRPT–at REM ADNM=sick

ʔə tsiʔəʔ ʔalš–s

PR PROX:FEM cross.sex.sibling–3PO

‘then he made a sickness so his sister got sick’

(Hess 1998: 56, line 16)

b. … ɬuacəxʷɬčiltxʷ dxʷʔal kʷsi adčəgʷas ʔi kʷi adbədbədaʔ

ɬu=ad=dəxʷ=ɬčil–txʷ dxʷ–ʔal kʷsi ad–čəgʷas

IRR=2SG.PO=ADNM–arrive–ECS CNTRPT–at REM:FEM 2SG.PO–wife

ʔi kʷi ad–bəd–bədaʔ

and REM 2SG.PO–DSTR–offspring

‘[you will pack it] so that you can bring it to your wife and children’

(Hess 1998: 86, line 237)

In such expressions, the adjunctive nominal is accompanied by the preposition dxʷʔal, whose

meaning ‘towards, motion to’ undoubtedly contributes to the purposive reading.

Even less frequent are adjunctive nominals that have a temporal reading:

(524) hay, gʷəl, diɬəxʷ tudəxʷkʷədatəbsəxʷ tiʔəʔ cədiɬ č’ətx%

hay gʷəl diɬ=əxʷ tu=dəxʷ=kʷəda–t–əb=s=əxʷ tiʔəʔ cədiɬ č’ətx%

INTJ then FOC=now PAST=ADNM=take–ICS–PASS=3PO=now PROX he kingfisher

‘well and then that is when they [almost] caught Kingfisher

(Hess 2006: 20, line 207)

The temporal aspect of such expressions may be as much as artefact of translation as the

semantics of the expression (cf. the alternate translations as locatives “ … where they caught

Kingfisher,” “… the point at which they caught Kingfisher”). Alternatively, the few uses of

dəxʷ= in temporal contexts may represent a metaphorical extension from the more common

400

locative spatial usage into time. There are no more than a handful of such expression in the texts

and so the temporal uses of adjunctive nominals appears to be a marginal function at best.

401

8 Predicative expressions

8.1 Grammatical relations

Like other Salishan languages, Lushootseed distinguishes three major argument types or

grammatical relations in a clause — the two “direct” or “nuclear” arguments, subject and direct

object, and a third argument type, the oblique object. The latter are realized by NPs or pronouns

introduced by the “empty” preposition ʔə, whereas the former — subject and direct object — are

realized either by means of bare noun phrases (that is, NPs introduced only by a determiner) or

by person-markers which distinguish between the grammatical relations of subjects and direct

objects as well as distinguishing different types of clauses. From a typological point of view

these person-markers are interesting in that they seem to be intermediate in their properties

between pronouns and agreement markers. Although they are clearly anaphora, they seem not to

be pronouns in that their distribution does not parallel that of nouns (as opposed to the

independent pronouns discussed in Section 2.7.1 above). Unlike agreement markers in most

languages, the Lushootseed person markers do not co-occur with overt NP or pronominal

arguments, as shown for the second-person singular object marker -bicid in the following

example:

(525) a. day’ čəxʷ ɬuʔəƛ’ dxʷʔal tiʔiɬ dʔalʔal čəda ɬuʔəɬtubicid

day’ čəxʷ ɬu=ʔəƛ’ dxʷ–ʔal tiʔiɬ d–ʔalʔal čəda

only 2SG.SUB IRR=come CNTRPT–at DIST 1SG.PO–house 1SG.COORD

ɬu=ʔəɬ–txʷ–bicid

IRR=eat–ECS–2SG.OBJ

‘after a while you will come to my house and I will feed you’

(Hess 1998)

b. ʔiˑ, ɬuk’ʷic’yid čəɬ ti dəgʷi

ʔi ɬu=k’ʷic’–yi–d čəɬ ti dəgʷi

INTJ IRR=butcher–DAT–ICS 1PL.SUB SPEC you

‘indeed, we will butcher it for you’

(Hess 1998)

402

By far the most typical manner of expressing the objects of transitive verbs is that illustrated in

(525a), where the object is expressed only by an object-marking suffix. In sentences with first- or

second-person objects represented by independent pronouns such as (525b), the corresponding

object-marker is absent. If object-markers are to be treated as agreement, they would have to be

treated as “deep” agreement with syntactic elements that are elided in the final surface form of

the sentence and which are in complementary distribution with overtly realized objects (cf. the

proposal in Hukari 1976). The same is true of subject-markers in matrix clauses and of first- and

second-person subject clitics in subjunctive subordinate clauses, although third-person subject

clitics in subjunctive subordinate clauses do co-occur with overt NP subjects — see example

(533c) below. Likewise, third-person subjects of nominalized clauses, expressed

morphologically by the third-person possessive suffix (Table 77), regularly co-occur with overt

NP subjects/possessors in certain constructions. First- and second-person possessive subjects co-

occurring with personal pronouns are not attested in the corpus to date.

As a result of these rather heterogeneous patterns, Lushootseed person-markers are (with a

few exceptions) in complementary rather than overlapping distribution with overt nominal and

pronominal arguments, and clauses with overt subjects and objects follow distinctly different

syntactic patterns than those containing subject- and object-markers. Because of this, the

different realizations of arguments will be dealt with separately, beginning with a discussion of

subject-markers in Section 8.1.1 and object-markers in Section 8.1.2 and then moving on to a

discussion of NP subjects and objects in Section 8.1.3. Section 8.1.6 ends the discussion of

grammatical relations with an examination of oblique objects.

8.1.1 Subject-markers

Subjects of both transitive and intransitive (non=nominalized) verbs in Lushootseed are

marked by one of five series of person-markers — matrix, co-ordinate, subjunctive, possessive,

403

or imperative — depending on the syntactic status of the clause. In matrix clauses, subject-

markers are drawn from the set of elements given in Table 79:

čəd

čəɬ

čəxʷ

čələp

Table 79: Matrix clause subject markers

Their use is exemplified in the following examples:

(526) a. ʔugʷəč’əd čəd ti sqʷəbayʔ

ʔu–gʷəč’–əd čəd ti sqʷəbayʔ

PFV–search–ICS 1SG.SUB SPEC dog

‘I looked for the dog’

b. ʔugʷəč’əd čəɬ ti sqʷəbayʔ

ʔu–gʷəč’–əd čəɬ ti sqʷəbayʔ

PFV–search–ICS 1PL.SUB SPEC dog

‘we looked for the dog’

c. ʔugʷəč’əd čəxʷ ti sqʷəbayʔ

ʔu–gʷəč’–əd čəxʷ ti sqʷəbayʔ

PFV–search–ICS 2SG.SUB SPEC dog

‘you

looked for the dog’

d. ʔugʷəč’əd čələp ti sqʷəbayʔ

ʔu–gʷəč’–əd čələp ti sqʷəbayʔ

PFV–search–ICS 2PL.SUB SPEC dog

‘you

looked for the dog’

e. ʔugʷəč’əd ti sqʷəbayʔ

ʔu–gʷəč’–əd Ø ti sqʷəbayʔ

PFV–search–ICS 3SUB SPEC dog

‘she/he/they looked for the dog’

(Hess 1995: 10, ex. a – e)

Any matrix clause lacking an overt subject marker (or an overt subject NP) is obligatorily

interpreted as having a third-person subject co-referent with the current discourse topic (Section

11.2.2), making both the third-person singular and plural matrix subject-marker an analytical

zero.

145

The neutralization of the singular/plural distinction in the third person is entirely

145

Unlike many other Salishan languages, Lushootseed has no special marker for third-person subjects in transitive

matrix clauses. Compare the Lummi example in (i) to the Lushootseed sentence in (526e):

404

consistent with the general Lushootseed treatment of nominal number. When is it considered

essential by the speaker to indicate the plurality of a third-person subject, the plural particle

əlgʷəʔ is used:

(527) gʷəx%aliǰəd əlgʷəʔ

gʷəx%–alič–əd əlgʷəʔ

unwrap–bundle–ICS PL

‘they unwrapped it’

(Hess 1998: 82, line 129)

The same particle is used optionally to express the plurality of objects (Section 8.1.2) and

subjects in other types of non-matrix clause.

The matrix subject-markers are used with all types of main clause, irrespective of the lexical

class of the predicate:

(528) a. xʷak’ʷiləxʷ čəd

xʷak’ʷil=əxʷ čəd

tired=now 1SG.SUB

‘I’m tired now’

(Hess 1995: 148, line 39)

b. ɬuk’ʷič’id čəɬ tiʔiɬ kʷagʷičəd

ɬu=k’ʷič’i–d čəɬ tiʔiɬ kʷagʷičəd

IRR=butcher–ICS 1PL.SUB this elk

‘we will butcher this elk’

(Hess 1998: 80, line 67)

Lummi

(i) x%či–t–s cə swəyʔqəʔ cə swiʔqoʔəɬ

know–TR–3 the man the boy

‘the man knows the boy’

(Jelinek & Demers 1983: 168)

In addition to the absence of a third-person subject suffix on the verb, Lushootseed also differs from Lummi in that

its syntax prohibits the realization of the subject NP in transitive clauses (see Section 8.2.2 for detailed discussion).

405

c. skəyu čəd

skəyu čəd

ghost 1SG.SUB

‘I’m a ghost’

(Hess 1998: 94, line 98)

In the first example here, the predicate of the sentence is an intransitive verb, xʷak’ʷiləxʷ ‘be

tired’, while in (528b) the predicate is a transitive verb, k’ʷič’id ‘butcher’, and in (c) the predicate

is a noun, skəyu ‘ghost, corpse’.

As in many Coast Salishan languages, the matrix subject-markers are sentence-second (S2)

particles. Thus, they typically follow the clausal predicate, as in (528), but may precede a verbal

predicate when the clause contains an adverbial or other qualifying element:

(529) cick’ʷ čəd ʔəxʷʔux%ʷəb

cick’ʷ čəd ʔəs–dxʷ–ʔux%ʷ–ab

very 1SG.SUB STAT–CTD–go–DSD

‘I very much want to go’

(Hess 1995: 90)

When another S2 particle appears in the same clause with a subject-marker, the subject marker

generally precedes it and both precede the predicate, although there are some exceptions. The

relative ordering of S2 particles will be discussed in more detail in Section 8.2.2.

The requirement for the matrix subject-marker to appear in sentence-second position also

applies to subjects of multi-word non-verbal predicates:

(530) a. tul’ʔal čəd sqaǰət

tul’ʔal čəd sqaǰət

from 1SG.SUB Skagit

‘I [am] from Skagit’ (Bates, Hess & Hilbert 1994: 6)

b. ɬuhikʷ čəd stubš ɬuluƛ’iləd

ɬu=hikʷ čəd stubš ɬu=luƛ’–il=əd

IRR=big 1SG.SUB man IRR=old–INCH=1SG.SBJ

‘I will be a big man when I grow old’

(Bates, Hess & Hilbert 1994: 109)

In example (530a), the first-person čəd appears inside the prepositional phrase, tul’ʔal sqaǰət

‘from Skagit’ while in (b) the person-marker separates the modifier hikʷ ‘big’ from its

406

modificand, stubš ‘man’, head of the nominal predicate, hikʷ stubš ‘big man’.

146

This is unusual

from a syntactic perspective in that the subject-marker interrupts the constituency of the

predicate phrase.

The second set of person-markers, the coordinate subject markers, is used in the second (and

subsequent) clauses of coordinate structures. These are given in Table 80:

čəd

čəɬ

čəxʷ

čələp

Table 80: Coordinate subject markers

While these are clearly derived diachronically from the matrix-subject markers in Table 79, the

formative suffix -a is not found in other parts of the grammar or used with coordinate NP

subjects, leading to the treatment of these elements as a separate series. The coordinate subject

markers are used in sentences such as those in (531):

(531) a. ƛ’ub čəɬ ʔuhudčup čɬa ʔukʷukʷcut

ƛ’ub čəɬ ʔu–hud•čup čɬa ʔu–kʷukʷcut

well 1PL.SUB PFV–wood•fire 1PL.COORD PFV–cook

‘we should make a fire and cook’

(Hess 1995: 114)

b. ɬuhuyud čəɬ čɬa ləsx%qaličtxʷ čxʷa ɬut’uk’ʷtxʷ

ɬu–huyu–d čəɬ čɬa ləs–x%q•alič–txʷ

IRR–be.done–ICS 1PL.SUB 1PL.COORD PROG.STAT–bound•bundle–ECS

čxʷa ɬu=t’uk’ʷ–txʷ

2SG.COORD IRR=go.home–ECS

‘we will fix it up and have it packaged and you will take it home’

(Hess 1998: 80, line 68)

Unlike the matrix subject-markers, the coordinate markers appear in front of their predicate, on

the left edge of the clause.

147

146

The placement of the irrealis prefix on the modifier, rather than the predicate head, is discussed in Section 3.1.

147

As noted by Hess (1993: 146), this may indicate that the previous coordinate in the sentence in some sense

“counts” as an element of the following clause, thereby requiring that the person-marker appear in its expected

second position, although this is a somewhat novel situation in that it implies that the rightmost — rather than the

leftmost — constituent is the head of the coordinate construction, in some sense “containing” the preceding clauses.

407

The third set of subject markers are used in certain types of subordinate clauses which are

commonly termed in Salishan studies conjunctive clauses (Thompson & Thompson 1992; Bates

1997a; Kroeber 1999). The term conjunctive, however, is somewhat unfortunate in that it implies

a coordinative function that these elements do not have, and which is in fact filled by the

coordinate subject-markers just discussed. Instead, these markers are associated with subordinate

clauses in a particular mood which corresponds (as also noted for Lushootseed by Hess and

Hilbert 1976 and for Salishan in general by Kroeber 1999) in some respects to the subjunctive

mood in Indo-European languages. Hence, I will instead adopt the term subjunctive

subordinate clause (see Section 9.3), and refer to these markers as subjunctive subject clitics.

These markers, while obviously etymologically related to the other two series, are quite distinct

on a number of counts, reflecting their origins in a different part of the Proto-Salishan person-

marking system than the matrix-subject series (Kroeber 1999; Davis 2000). The most obvious

difference is that the subjunctive person-markers are bound enclitics rather than S2 particles, and

each has two allomorphs; these are given in Table 81:

148

=ad / =əd

=aɬi / =əɬi

=axʷ / =əxʷ

=aləp / =ələp

=as / =əs

Table 81: Subjunctive subject clitics

The a-allomorphs occur when the clitic receives stress — that is, in those cases where they

appear on a stem containing only schwa vowels:

(532) a. ɬuʔəɬtxʷas ti ʔaciɬtalbixʷ

ɬu=ʔəɬ–txʷ=as ti ʔaciɬtalbixʷ

IRR=feed–ECS=3SBJ SPEC person

‘when/if she feeds the people’

148

The first-person plural clitic given in Table 81 is a Northern Lushootseed form; in Southern Lushootseed the

form is =aɬ čəɬ.

408

b. ɬuʔəɬtubəs ti ʔaciɬtalbixʷ

ɬu=ʔəɬ–txʷ–b=əs ti ʔaciɬtalbixʷ

IRR=feed–ECS–PASS=3SBJ SPEC person

‘when/if the people feed her’

(Hess 1995: 109)

The schwa forms occur when the clitic is added to a stem containing a non-schwa vowel (in

which case stress falls on the leftmost non-schwa vowel in the stem), as shown in (533):

(533) a. ƛ’uhiiɬ čəɬ ɬuʔukʷukʷəɬi

ƛ’u=hiiɬ čəɬ ɬu=ʔukʷukʷ=əɬi

IRR=happy 1PL.SUB IRR=play=1PL.SBJ

‘we are usually happy when we play’

b. ʔut’iwiɬtxʷ čəd ɬuq’ilidəs

ʔu–t’iwiɬtxʷ čəd ɬu=q’ili–d=əs

PFV–thank 1SG.SUB IRR=aboard–ICS=3SBJ

‘I asked him to give her a ride’

(Hess 1995: 71, ex. 3 – 4)

c. x%ʷul’ čəd gʷəqʷiqʷəbqʷəbayʔcut gʷəʔəƛ’as kʷi tubšədəʔ

x%ʷul’ čəd gʷə=qʷi–qʷəb–qʷəbayʔ–t–sut gʷə=ʔəƛ’=as kʷi tubšədəʔ

just 1SG.SUB SBJ=ATTN–DSTR–bark–ICS–REFL SBJ=come=3SBJ REM warrior

‘I would just make myself sound like a dog if the warriors came’

(Bates 1997a: 320)

d. ʔəsx%əc gʷəxʷit’iləs əlgʷəʔ

ʔəs–x%əc gʷə=xʷit’il–əs əlgʷəʔ

STAT–afraid SBJ=fall–3SBJ PL

‘he is afraid they will fall’

(Hess 1967a: 76)

The subjunctive person-marker series differs from the matrix and coordinate series in that it has

an overt third-person form (533b) which co-occurs with subject NPs (533c). As with matrix

subjects, plural number for third-persons is optionally marked with the plural particle əlgʷəʔ

(533d).

Like the matrix person-markers, the subjunctive clitics gravitate to second position in the

sentence; however, unlike sentence-second particles, the subjunctive person-markers are bound

enclitics that become phonologically dependent on the immediately preceding word:

409

(534) a. gʷəck’ʷaqidələp gʷučaləc

gʷə=ck’ʷaqid=ələp gʷə–ʔu–čala–t–s

SBJ=always=2PL.SBJ SBJ=PFV–chased–ICS–1OBJ

if you folks always chase me’

(Hess 1967a: 52)

b. tux%ʷəxʷ stab gʷəl ʔaləs tadiʔ siq’gʷas ʔə tə šəgʷɬ

tux%ʷ=əxʷ stab gʷəl ʔal=əs tadiʔ s–siq’•gʷas ʔə tə šəgʷɬ

just=now what SCONJ at=3SBJ DIST.DMA NP–spread•pair PR NSPEC path

‘(they) are just where there is a fork in the path over there’

[AW Basket Ogress, line 99]

In the first example here, the subjunctive person marker appears in a clause introduced by an

adverb, ck’ʷaqid ‘always’; because the adverb is clause-initial, the person-marker appears

encliticized to this word rather than to the verb, thus maintaining sentence-second position.

Similarly, the sentence in (534b) illustrates the same phenomenon with a clause predicated on a

prepositional phrase. In this case, the person-clitic appears bound to the preposition, the initial

element in the PP, creating a discontinuous constituent.

As shown in (535), when the person-markers co-occur with the bound enclitic =axʷ (Section

3.1.1), =axʷ precedes the subjunctive person-marker.

(535) a. huy xʷuʔələʔ cutəbəxʷ ʔə tiʔəʔ təkʷtekʷəlus tsiʔəʔ čəgʷas gʷəƛ’uhaʔliʔɬəxʷəs

huy xʷuʔələʔ cut–t–əb–əxʷ ʔə tiʔəʔ təkʷtekʷəlus tsiʔəʔ

SCONJ maybe speak–ICS–PASS–now PR PROX owl PROX:FEM

čəgʷas–s gʷə=ƛ’u=haʔl•iʔɬ=əxʷ=əs

wife–3PO SBJ=HAB=stop.crying•child=now=3SBJ

‘then Owl must have told his wife that she should tend to the baby.’

(Hess 2006: 4, line 24)

b. … gʷəʔux%ʷəxʷəɬi

gʷə=ʔux%ʷ=əxʷ=əɬi

SBJ=go=now=1PL.SBJ

‘… if we go now’

(Hess 1995: 70)

Combined with the attraction of the subjunctive person-markers to sentence-second position,

this is good evidence that they are best considered clitics associated with the predication at the

clausal level rather than as word-level markers of inflectional agreement.

410

The fourth series of subject-markers are homophonous with the possessive markers

introduced in Table 77 in Section 7.2 above:

čəɬ

ad=

=ləp

Table 82: Possessive subject-clitics

These markers are used for the subjects of nominalized clauses such as those in (536):

(536) a. tul’čad kʷi adsuʔibəš, sgʷəlub

tul’–čad kʷi ad=s=ʔu–ʔibəš sgʷəlub

from–where REM 2SG.PO=NM–PFV–travel pheasant

‘Where are you traveling from, Pheasant?’

(Hess 1998: 79, line 43)

b. hikʷ kʷagʷičəd tiʔiɬ səsɬild čəɬ ti dəgʷi

hikʷ kʷagʷičəd tiʔiɬ s=ʔəs–ɬil–d čəɬ ti dəgʷi

big elk DIST NM=STAT–give.food–ICS 1PL.PO SPEC you

‘our gift of food to you [is] a big elk’

(Hess 1998: 81, line 110)

c. tiʔiɬ ʔalʔals ƛ’udəxʷʔux%ʷs ʔal tiʔiɬ pədt’əs

tiʔiɬ ʔalʔals ƛ’u=dəxʷ=ʔux%ʷ=s ʔal tiʔiɬ pədt’əs

DIST house–3PO HAB=NM=go=3PO PR DIST winter

‘his house where he would go in the winter’

(Hess 1995: 143, line 7)

d. tul’čad kʷi skʷədxʷs əlgʷəʔ

tul’čad kʷi s=kʷəd–dxʷ=s əlgʷəʔ

from–where REM NM=take–DC=3PO PL

‘from where did they manage to get it?’

(Hess 1998: 83, line 160)

As seen in these examples, the possessive subject markers — including the first-person plural

possessive clitic čəɬ borrowed from the matrix-subject series (536b) — are used for the subjects

of clauses bearing the nominalizing proclitics s= (536a – b) and dəxʷ= (536c). When it is

essential to mark the plurality of the third-person subject, the plural particle əlgʷəʔ is used, as it

is when marking other types of third-person plural subjects (see 527 and 533d above) and

possessors (457).

411

The fact that the possessive subject markers are S2 clitics rather than affixes is shown by the

examples in (537) (see also the examples in 511 above):

(537) tux%ʷ čəɬ x%aƛ’txʷ kʷi sƛ’ubləp ʔəsqʷib …

tux%ʷ čəɬ x%aƛ’–txʷ kʷi s=ƛ’ub=ləp ʔəs–qʷib

PTCL 1PL.SUB desire–ECS REM NM=well=2PL.PO STAT–prepared

‘we want you guys to be well-prepared …’

(Hess 2006: 74, line 759)

In this example, the clause ƛ’ub čələp ʔəsqʷib ‘you guys are well-prepared’ appears as a

subordinate clause nominalized by the proclitic s=. As a result of being nominalized, it requires

the possessive subject-marking enclitic =ləp, which appears (along with the nominalizing

proclitic) on the first element in the subordinate clause, the adverb ƛ’ub ‘well’, rather than on the

head the verb phrase, qʷib ‘prepared’. As with the nominalizer s= and the subjunctive proclitic

gʷə=, the possessive subject-markers are another example of morphemes that have parallels on

both the lexical and the phrasal level. Further evidence that the possessive subject markers are

clitics can also be seen in the variable ordering of the third-person clitic =s relative to the bound

enclitic =axʷ, discussed in Section *.*.

As noted earlier in Section 7.4.2, third-person subjects of nominalized clause have the same

range of possible expressions as do third-person possessors:

(538) a. ɬuɬəgʷɬ čəɬ ʔal ti ƛ’usəsʔituts

ɬu=ɬəgʷɬ čəɬ ʔal ti ƛ’u=s–ʔəs–ʔitut=s

IRR=leave 1PL.SUB at SPEC HAB=NM–STAT–sleep=3PO

‘we will leave while he sleeps’

(Hess 1998: 101, line 271)

b. diɬ day’ ƛ’uscut ʔə tiʔiɬ sčətxʷəd

diɬ day’ ƛ’u=s=cut ʔə tiʔiɬ sčətxʷəd

FOC only HAB=NM=say PR DIST bear

‘that was Black Bear’s habitual saying’

(Hess 1995: 145, line 47)

(538a) shows the subject realized as the third-person possessive suffix, -s, while in (538b) the

subject is an overt NP, tiʔiɬ sčətxʷəd ‘that black bear’ is expressed by a periphrastic possessive

412

construction (cf. (456) above). This is the preferred form for the expression of the subject of a

nominalized clause; however, the possessive subject clitics can co-occur with overt third-person

subjects, as in(539) (cf. 455):

(539) a. stab kʷi suhuyitəbs ti adbad tiʔiɬ č’ač’as

stab kʷi s=ʔu–huy–yi–t–əb=s ti ad–bad tiʔiɬ č’ač’as

what REM NM=PFV–be.done–DAT–ICS–PASS=3PO SPEC 2SG.PO–father DIST child

‘what is your father making for that boy?’

(Hess 1995: 106)

149

b. diɬəxʷ tiʔiɬ dəxʷəsluutəbs tsiʔiɬ cədiɬ diver, x%ʷuʔx%ʷəyʔ

diɬ=əxʷ tiʔiɬ dəxʷ=ʔəs–lu–t–əb=s tsiʔiɬ cədiɬ

FOC=now DIST ADNM=STAT–be.heard–ICS–PASS=3PO DIST:FEM her

diver x%ʷuʔx%ʷəyʔ

diver helldiver

‘that is the way that Helldiver sounds’

(Hess 2006: 13, line 57)

The pattern in (539a) seems to be the required one for nominalized passive constructions, and it

seems to be limited largely to these, at least for s=nominals. Clausal nominals formed with

dəxʷ=, however, can take this form even in the active voice:

(540) gʷəl təɬ biƛ’il tiʔiɬ tudəxʷux%aabsəxʷ tsiʔiɬ cədiɬ sʔušəbabdxʷ

gʷəl təɬ biƛ’–il tiʔiɬ tu=dəxʷ=ʔu–x%aab=s=əxʷ tsiʔiɬ cədiɬ

then truly crushed–INCH DIST PAST=ADNM=PFV–cry=3PO DIST:FEM she

sʔušəbabdxʷ

poor.dear

‘then he truly got squashed, which is why the poor dear was crying’

(Hess 1998: 72, line 173)

This pattern may represent something of an archaism, given that the same pattern in ordinary

possessive constructions has been supplanted by the periphrastic construction in (456) and

(539b).

149

The analysis of this sentence presented here differs from that presented in Hess (1993), which analyzes the third-

person possessive suffix -s on the verb as cross-referencing the clefted oblique object, stab, which is the THEME of

the embedded clause. However, since the direct object of an applicative verb such as huyid ‘make something for

someone’ is the RECIPIENT rather than the THEME, we would expect the RECIPIENT, tiʔiɬ č’ač’as ‘the boy’, to be the

subject of the passive — and, therefore, the argument cross-referenced by the possessive subject-marker.

413

The final subject paradigm is the imperative subject paradigm, which has values only for the

second person:

ɬi

Table 83: Imperative subject markers

This paradigm presents a simple contrast between a zero singular and an overt plural, as shown

in the sentences in (541):

(541) a. lilcut

lil–t–sut Ø

far–ICS–REFL 2SG.IMP

‘go away!’

(Hess 1998: 83, line 152)

b. lilcut ɬi

lil–t–sut ɬi

far–ICS–REFL 2PL.IMP

‘go away, you guys!’

(Hess 1998: 93, line 60)

Because the singular member of the paradigm is zero and second-person plural imperatives are

textually infrequent (and possibly pragmatically dispreferred), little is currently known about the

syntactic behaviour of the imperative subject clitics in the presence of preverbal elements such as

adverbs and adverbial particles. It seems likely that they mimic the other subject-markers and

would migrate to sentence-second position, but this will have to remain in the realm of

conjecture until further examples are uncovered. Imperative constructions are discussed in detail

in Section 8.5.

In summary, the five series of subject-markers are shown together in Table 84:

MATRIX

COORDINATE

SUBJUNCTIVE

POSSESSIVE

IMPERATIVE

1SG

čəd

čəda

=ad

—

1PL

čəɬ

čɬa

=aɬ

čəɬ

—

2SG

čəxʷ

čxʷa

=axʷ

ad=

2PL

čələp

čələpa

=aləp

=ləp

ɬi

=as

—

Table 84: Summary table of subject-markers

414

These markers are all clearly etymologically related and they have obvious cognates in other

languages of the family. The history of the subject-markers in Salishan languages is discussed in

a number of places in the literature, notably Newman (1969; 1977; 1979; 1980), Kroeber (1999),

and Davis (2000).

8.1.2 Object-markers

Transitive verbs in Lushootseed are inflected for the person and (in first- and second-person)

number of their direct objects. The object markers can be divided into two series, the s-series and

the b-series,

150

one associated with verb stems formed with the [-t], [-d], and [-əd] allomorphs of

the internal causative suffix, the other associated with stems formed with the other valency-

increasing morphemes, -txʷ, -dxʷ, and -s/-c, as well as the [-š] allomorph of the internal

causative. The two series of object markers are given in Table 85:

s-series

b-series

1SG

-s

-bš

2SG

-sid

-bicid

1PL

-ubuɬ

-buɬ

2PL

-ubuɬəd

-buɬəd

Table 85: Object-markers

Example forms for s-series object-markers are given in (542):

(542) a. ʔukʷədac b. ʔukʷədacid

ʔu–kʷəda–t–s ʔu–kʷəda–t–sid

PFV–grab–ICS–1SG.OBJ PFV–grab–ICS–2SG.OBJ

‘s/he grabbed me’ ‘s/he grabbed you’

c. ʔukʷədatubuɬ d. ʔukʷədatubuɬəd

ʔu–kʷəda–t–ubuɬ ʔu–kʷəda–t–ubuɬəd

PFV–grab–ICS–1PL.OBJ PFV–grab–ICS–2PL.OBJ

‘s/he grabbed us’ ‘s/he grabbed you guys’

e. ʔukʷədad f. ʔukʷədad čəd

ʔu–kʷəda–t–Ø Ø ʔu–kʷəda–t–Ø čəd

PFV–grab–ICS–3SG.OBJ 3SUB PFV–grab–ICS–3SG.OBJ 1SG.SUB

‘she/he/they grabbed him/her/it/them’ ‘I grabbed him/her/it/them’

150

Cf. Montler’s (1996) S-object versus M-object distinction in Klallam.

415

As seen in (542a) and (b), the first- and second-person markers undergo phonological fusion

with the transitivizing suffix whereby a /t + s/ sequence becomes a single phoneme, /c/.

The b-series object-markers also interact with the final segment of their stems, the results in

this case depending on what exactly that final segment is. With the allative applicative, -s/-c, b-

series object-markers are accompanied most cases by schwa-epenthesis:

(543) a. ʔutəlawisəbš b. ʔutəlawisəbicid

ʔu–təlawil–s–bš ʔu–təlawil–s–bicid

PFV–run–ALTV–1SG.OBJ PFV–run–ALTV–2SG.OBJ

‘s/he ran after me’ ‘s/he ran after you’

c. ʔutəlawisəbuɬ d. ʔutəlawisəbuɬəd

ʔu–təlawil–s–buɬ ʔu–təlawil–s–buɬəd

PFV–run–ALTV–1PL.OBJ PFV–run–ALTV–2PL.OBJ

‘s/he ran after us’ ‘s/he ran after you guys’

e. ʔutəlawis f. ʔutəlawis čəd

ʔu–təlawil–s–Ø Ø ʔu–təlawil–s–Ø čəd

PFV–run–ALTV–3SG.OBJ 3SUB PFV–run–ALTV–3SG.OBJ 1SG.SUB

‘she/he/they ran after him/her/it/them’ ‘I ran after him/her/it/them’

When the stems ends in -txʷ or -dxʷ, the final consonant of the stem (that is, /xʷ/) becomes /u/:

(544) a. ʔubəčdubš b. ʔubəčdubicid

ʔu–bəč–dxʷ–bš ʔu–kʷəda–dxʷ–bicid

PFV–fall–DC–1SG.OBJ PFV–grab–DC–2SG.OBJ

‘s/he knocked me down’ ‘s/he knocked you down’

c. ʔubəčdubuɬ d. ʔubəčdubuɬəd

ʔu–bəč–dxʷ–buɬ ʔu–bəč–dxʷ–buɬəd

PFV–fall–DC–1PL.OBJ PFV–fall–DC–2PL.OBJ

‘s/he knocked us down’ ‘s/he knocked you guys down’

e. ʔubəčdxʷ f. ʔubəčdxʷ čəd

ʔu–bəč–dxʷ–Ø Ø ʔu–bəč–dxʷ–Ø čəd

PFV–fall–DC–3SG.OBJ 3SUB PFV–fall–DC–3SG.OBJ 1SG.SUB

‘she/he/they knocked him/her/it/them down’ ‘I knocked him/her/it/them down’

Because of the interaction of the final /xʷ/ with the first- and second-person plural object-

markers, these affixes become homophonous with the corresponding forms in the s-series.

416

Stems formed with the [-š] allomorph of the internal causative also interact

morphophonemically with object-markers, /-š/ becoming /i/ and taking a marker from the b-

series, as in the following examples:

(545) a. ʔuɬalš čəɬ tsi kiaʔləp

ʔu–ɬal–š čəɬ tsi kiaʔ–ləp

PFV–remove.from.fire–ICS 1PL.SUB SPEC:FEM grandmother–2PL.PO

‘we took your grandmother out of the fire’

[AJ Basket Ogress, line 105]

b. ɬalibš ɬi wiw’su

ɬal–š–bš ɬi wiw’su

remove.from.fire–ICS–1SG.OBJ 2PL.IMP children

‘get me out of the fire, children!’

[AJ Basket Ogress, line 101]

For those stems with both an -š and a -d variant (Table 21), the b-series object forms are only

found on the stem formed with the [-š] allomorph, while the other form takes regular s-series

object-markers.

Unlike the subject-markers (discussed in Section 8.1.1 above), the object-markers are

compatible with and invariant for all clause-types. Thus, they appear in the forms given in Table

85 in matrix clauses,

(546) a. ɬut’uk’ʷtubicid čəd

ɬu–t’uk’ʷ–txʷ–bicid čəd

IRR–go.home–ECS–2SG.OBJ 1SG.SUB

‘I will take you home’

b. ɬuʔux%ʷtubuɬ čələp ʔu

ɬu=ʔux%ʷ–txʷ–buɬ čələp ʔu

IRR=go–ECS–1PL.OBJ 2PL.SUB INT

‘will you guys take us?’

c. ɬuʔux%ʷtubš čələp ʔu

ɬu=ʔux%ʷ–txʷ–bs čələp ʔu

IRR=go–ECS–1SG.OBJ 2PL.SUB INT

‘will you guys take me?’

417

d. ɬut’uk’ʷtubuɬəd čəɬ

ɬu=t’uk’ʷ–txʷ–buɬəd čəɬ

IRR=go.home–ECS–2PL.OBJ 1PL.SUB

‘we will take you guys home’

(Hess 1995: 41, ex. 5 – 8)

as well as in subjunctive subordinate clauses (9.3),

(547) a. ʔəsx%əc čəd gʷəlax%dubšəs

ʔəs–x%əc čəd gʷə=lax%–du–bš=əs

STAT–afraid 1SG.SUB SBJ=remember–DC–1SG.OBJ=3SG.SBJ

‘I’m afraid he might remember me’

(Bates, Hess & Hilbert 1994)

b. ƛ’ub dxʷʔal ʔəca gʷəgʷabicəxʷ

ƛ’ub Ø dxʷʔal ʔəca gʷə=gʷah–bi–t–s=əxʷ

good 3SG.SUB for me SBJ=accompany–MAP–ICS–1SG.OBJ=2SG.SBJ

‘it’s okay with me if you accompany me’

(Hess & Hilbert 1976: II, 39)

and nominalized clauses (548),

(548) a. xʷiʔ kʷi ɬadswəliʔtubuɬ

xʷiʔ kʷi ɬu=ad=s=wəliʔ–t–ubuɬ

NEG REM IRR=2SG.PO=NM=reveal–ICS–1PL.OBJ

‘you will not reveal us’

(Bierwert 1996)

b. … ɬuswəliʔcid čəɬ dxʷʔal tiʔiɬ adyəl’yelab

ɬu=s=wəliʔ–t–sid čəɬ dxʷ–ʔal tiʔiɬ ad–yəl’–yelab

IRR=NM=reveal–ICS–2SG.OBJ 1PL.PO CNTRPT–at DIST 2SG.PO–PL–forebear

‘our revealing of you to your older relatives’

(Bierwert 1996: 189, line 130)

c. … čad ʔal kʷi ɬudsɬčisəbuɬəd

čad ʔal kʷi ɬu=d=s=ɬčil–s–buɬəd

where at REM IRR=1SG.PO=NM=arrive–ALTV–2SG.OBJ

‘where I will go to for you guys’

(Hess 2006: 68, line 627)

In addition to being used in all types of clause, the object-markers are freely combinable with all

logically possible persons and numbers of subjects, as illustrated in the examples given above.

This makes Lushootseed different from a number of other closely-related languages which have

418

restrictions — particularly with second-persons — as to permissible combinations of person and

number of subjects and objects in transitive clauses (Jelinek & Demers 1983).

As with subject- and possessive-markers, objects do not obligatorily distinguish number in

the third person. When it is considered necessary to do so by the speaker, the plural particle

əlgʷəʔ is used:

(549) ʔəšudxʷ čəd əlgʷəʔ

ʔəs–šuɬ–dxʷ čəd əlgʷəʔ

STAT–see-DC 1SG.SUB PL

‘I see them’

(Hess & Hilbert 1976: II, p. 87)

The fact that the plural particle can serve equally to pluralize objects, subjects, and possessors

argues against its inclusion as a member of any one in particular of these person-paradigms.

8.1.3 Reflexive-markers

Like the transitive object-markers, the reflexive marker is attached to transitivized stems and

has two forms, -sut and -but, which are distributed according to the final element of the stem to

which they are attached. For words ending in the transitivizer -t, the form is -sut, as in (550):

(550) a. gʷəl ləq’ʷuʔcut ʔə tiʔəʔ caadiɬ

gʷəl lə=q’ʷuʔ–t–sut ʔə tiʔəʔ caadiɬ

then PROG=gathered–ICS–REFL Pr this they

‘then he joined them’ (lit. ‘he gathers himself to them’)

(Hess 1998: 79, line 41)

b. ʔut’uc’ucut čəxʷ ʔu

ʔu–t’uc’u–t–sut čəxʷ ʔu

PFV–be.shot–ICS–REFL 2SG.SUB INT

‘did you shoot yourself?’

(Hess 1995: 43, ex. 11b)

Because of the fusion of the reflexive marker with the internal causative suffix, the reflexive

maker surfaces as [cut]. As expected, radicals such as √t’uc’ ‘be shot’ in (550b) undergo the

same alternations in the reflexive as in the active form — thus, the internal causative form of

√t’uc’ is t’uc’ud ‘shoot something’ and the reflexive form is t’uc’ucut ‘shoot oneself’. There is,

419

however, one stem — ǰiq’cut ‘soak oneself’ (cf. ǰiq’id ‘soak something’) — which is attested

without the harmonic vowel.

Words ending in the transitivizers other than the internal causative take the -but form of the

suffix:

151

(551) ʔuʔəɬtubut čəd

ʔu–ʔəɬ–txʷ–but čəd

PFV–feed.on–ECS–REFL 1SG.SUB

‘I fed myself’

(Hess 1995: 43, ex. 10c)

As is to be expected, the application of the reflexive marker to a verb stem renders that stem

intransitive, precluding the appearance of object-markers or the realization of a direct object.

In addition to its regular use as a true reflexive, the reflexive morpheme appears in a few

idiomatic expressions affixed to bases that are not attested on their own as transitive stems. Two

of these are given in (552):

(552) a. tuwəliʔil tiʔəʔ swatixʷtəd tiʔiɬ, tə tudəxʷʔəy’cuts

tu=wəliʔ–il tiʔəʔ swatixʷtəd tiʔiɬ tə tu=dəxʷ=ʔəy’–t–sut=s

PAST=visible–INCH PROX country PROX SPEC PAST=ADNM=found–ICS–REFL=3PO

‘the land there, where they found themselves, became visible’

(Hess 2006: 75, line 779)

b. huy, dᶻəgʷaʔcut

huy dᶻəgʷaʔ–t–sut

SCONJ expert–ICS–REFL

‘then he did something clever’

[DM Basket Ogress, line 20]

The form in (552a), ʔəy’cut ‘find oneself in a place’, appears to be the reflexive of the unattested

form *ʔəy’əd ‘find something’ (from the radical º√ʔəy’ ‘be found’; cf. ʔəy’dxʷ ‘find something’).

Although it calques nicely into idiomatic English, ʔəy’cut is not a transparent use of the reflexive

and needs to be treated as a phraseologized form. Similarly, the verb dᶻəgʷaʔcut ‘do something

151

Reflexive forms of stems ending in the allative applicative -c/-s are unattested (Hess 1995: 43), but are presumed

to follow this pattern in that they group with the -txʷ and -dxʷ transitivizers in taking the b-series object-markers.

Note, however, that Gerdts (1988) has found that reflexives are not possible at all with the analogous applicative in

Halkomelem.

420

clever, get smart’ (cf. Spanish ponerse listo) represents a phraseologized use of a noun, dᶻəgʷaʔ

‘expert’, which would not normally be expected to take the internal causative suffix -t. A list of

other reflexive forms without corresponding transitives is given in :

ʔaɬcut ‘hurry’

(√ʔaɬ ‘be fast’)

ʔəy’cut ‘find oneself in a place’

(º√ʔəy’ ‘be found’; cf. ʔəy’dxʷ ‘find something’)

čaɬaʔcut ‘distance oneself from ’

(º√čaɬaʔ ‘not be recognized’; cf. čaɬaʔdxʷ ‘not recognize ’)

č’itcut ‘come close, approach’

(√č’it ‘nearby’)

ləliʔcut ‘change oneself’

(√ləliʔ ‘be different’)

p’alilcut ‘come to one’s senses’

(√p’alil ‘revive’)

q’iƛ’cut ‘injure oneself’

(√q’iƛ’ ‘be wounded’)

x&ix&q’cut ‘talk back, argue’

(√x&ix&q’ ‘insist’)

x&ʷax&ʷaq’ʷcut ‘be troubled, be worried’

(√x&ʷax&ʷaq’ʷ ‘feel concern’)

Table 86: Reflexive verbs without plain transitive forms

As noted in Hess (1967a: 22–23), there are also few stems such as ƛ’əlabut ‘understand, be able

to hear’, ləqalbut ‘understand, be able to hear’, šuɬalbut ‘be able to see’, ʔitut ‘sleep’, qʷəscut

‘approach stealthily fluttering wings’, and xʷalitut ‘snore’, that appear to contain the reflexive

marker but which are not synchronically analyzable and which do not seem to contain a valency-

increasing affix. Similarly, there are a few forms such as daʔcutbid ‘tell someone one’s own

name’ (based on √daʔ ‘be named’) and x&alcutxʷ ‘take a picture’ (from √x&al ‘be written’) in

which the reflexive form of the verb appears to have been lexicalized as the base for subsequent

derivation. These forms present a bit of an etymological mystery, but are beyond the scope of a

synchronic grammar.

8.1.4 Reciprocal-markers

The reciprocal suffix -agʷəl ‘reciprocal [RCP]’ follows transitive stems and occupies the slot

normally occupied by the object-markers:

421

(553) a. ʔuqʷulutagʷəl

ʔu–qʷulu–t–agʷəl

PFV–hugged–ICS–RCP

‘they hugged one another’

(Hess 1995: 43, ex. 16a)

b. ʔuɬəčisagʷəl

ʔu–ɬəčil–s–agʷəl

PFV–arrive–ALTV–RCP

‘they come together’

(Hess 1995: 43, ex. 17)

When the reciprocal follows one of the two causatives ending in /xʷ/, this segment is elided:

(554) a. ʔuʔəɬtagʷəl čəɬ

ʔu–ʔəɬ–txʷ–agʷəl čəɬ

PFV–feed.on–ECS–RCP 1PL.SUB

‘we fed one another’

(Hess 1995: 43, ex. 13a)

b. ʔušudagʷəl čəɬ

ʔu–šuɬ–dxʷ–agʷəl čəɬ

PFV–see–DC–RCP 1PL.SUB

‘we saw one another’

(Hess 1995: 43, ex. 14a)

Like the reflexive marker, the reciprocal intransitivizes the verb stem to which it is attached.

Except in those cases where the verb stem contains no non-schwa vowels, the reciprocal marker

is most frequently realized phonetically as [əgʷəl] due to normal processes of unstressed vowel

reduction, although this reduction is not always recognized in transcription.

8.1.5 Subject, object, and direct complement noun phrases

When subjects and objects are not expressed by person-markers, they are expressed either by

pronouns (Section 2.7.1) or full NPs (Section 7). Because Lushootseed lacks nominal case, the

form of subject and object phrases is identical:

(555) a. hay qɬaxʷ tiʔiɬ sčətxʷəd

hay qɬ=axʷ tiʔiɬ sčətxʷəd

INTJ awaken=now DIST bear

‘well, Black Bear awoke’

(Hess 1995: 154, line 69)

422

b. hay gʷəl tulax%dxʷəxʷ tiʔiɬ syaʔyaʔs

hay gʷəl tu=lax%–dxʷ=əxʷ tiʔiɬ syaʔyaʔ–s

INTJ SCONJ PAST=remember–DC=now DIST relative–3PO

‘and then he remembered his relative’

(Hess 1995: 151, line 2)

c. hikʷ kʷagʷičəd tiʔiɬ səsɬild čəɬ ti dəgʷi

hikʷ kʷagʷičəd tiʔiɬ s=ʔəs–ɬil–d čəɬ ti dəgʷi

big elk DIST NM=stat–give.food–ICS 1PL.PO SPEC you

‘what we give to you [is] a big elk’

(Hess 1998: 81, line 110)

The normal position for subject and object NPs in Lushootseed is immediately following the

sentence predicate, whether the predicate is a verb as in (555a) and (b) or an NP (555c). As

discussed in Section 7.3 above, complex expression such as the headless relative clause in (555c)

can and frequently do serve as subjects and objects. Arguments other than subjects and objects

are expressed as oblique objects introduced by a preposition (Section 8.1.6).

A significant fact about clause structure in Lushootseed is the fact that a grammatical

sentence can not contain more than one non-oblique NP. This pattern is also seen on other

Salishan languages, though in less absolute form. As a result, the term “direct complement” or

“direct argument” is frequently employed in Salishan studies to refer to the single non-oblique

NP in a clause. This term has proven useful in discussions of the valency and orientation of verb-

stems (see Section *.*) and in the discussion of such process as relativization and clefting, which

treat subjects and objects on a par and treat them differently from oblique objects. However, in

the context of a discussion of grammatical relations, the term direct complement can be

confusing, as it applies equally to all of the NPs in the following examples:

(556) a. ʔibəšəxʷ tiʔəʔ sgʷəlub

ʔibəš=əxʷ tiʔəʔ sgʷəlub

travel=now PROX pheasant

‘Pheasant travels

(Hess 1998: 78, line 11)

423

b. ʔuʔux%ʷtxʷ tsi č’ač’as

ʔu–ʔux%ʷ–txʷ tsi č’ač’as

PFV–go–ECS SPEC:FEM child

‘[he/she/they] took the girl’

(Hess 1995: 33)

c. ʔugʷəč’əb ti luƛ’ ʔə ti sqəlalitut

ʔu–gʷəč’–əb ti luƛ’ ʔə ti sqəlalitut

PFV–look.for–CSMD SPEC elder PR SPEC guardian.spirit

‘the old man looked for a guardian spirit’

(Hess 1995: 32)

As shown in (556), the term direct complement applies to the subject of a monovalent

intransitive verb (a), the direct object of a transitive verb (b), and the subject of a bivalent

intransitive verb (c). If the direct complement is replaced by a person-marker, the choice of

paradigm is different in each case:

(557) a. tux%ʷ čəd ƛ’uʔibəš

tux%ʷ čəd ƛ’u=ʔibəš

just 1SG.SUB HAB=travel

‘I’m just walking around’

(Hess 1998: 79, line 44)

b. ʔuʔux%ʷtubš

ʔu–ʔux%ʷ–txʷ–bš

PFV–go–ECS–1SG.OBJ

‘[he/she/they] took me’

(Hess 1995: 42)

c. ʔugʷəč’əb čəd ʔə ti sqəlalitut

ʔu–gʷəč’–b čəd ʔə ti sqəlalitut

PFV–look.for–CSMD 1SG.SUB PR SPEC guardian.spirit

‘I looked for a guardian spirit’

(Hess 1995: 32)

The direct complements in (557a) and (c) correspond to subject pronouns, while the direct

complement in (b) corresponds to an object pronoun. Furthermore, the object pronoun in (557b)

is compatible with both a matrix subject-marker (558a) and an NP (b), both of which express the

agent of the event (as opposed to the NP in (556b), which expresses a patient):

424

(558) a. ɬuʔux%ʷtubš ʔu čələp

ɬu=ʔux%ʷ–txʷ–bš ʔu čələp

IRR=go–ECS–1SG.OBJ INT 2PL.SUB

‘will you guys take me?’

(Hess 1995: 41, ex. 7)

b. ʔuʔux%ʷtubš tsi luƛ’

ʔu–ʔux%ʷ–txʷ–bš tsi luƛ’

PFV–go–ECS–1SG.OBJ SPEC:FEM elder

‘the old woman took me’

(based on Hess 1995: 41, ex. 2b)

Thus, in the presence of an object-marker, the direct complement of a patient-oriented stem like

ʔux&ʷtxʷ ‘take’ corresponds to a subject-marker, whereas in the absence of an object-marker is

corresponds to an object-marker. It seems clear from these examples that the term direct

complement does not refer to a grammatical relation but is instead a cover term for a non-oblique

NP. The grammatical relation of this NP to the verb depends on the transitivity of the verb and

the presence or absence of overt person-markers in the clause. Given that both third-person

subject-markers and third-person object-markers are zero, this gives rise to a potential ambiguity

in transitive clauses with a third-person direct complement such as (556b) (which could have

either a zero object-marker or a zero subject-marker). This ambiguity in Lushootseed and other

Salishan languages is resolved by a rule of conventional interpretation, the “One=nominal

Interpretation” rule (Gerdts 1988), whereby the single non-oblique NP argument is always

interpreted as the patient/direct object of the expression.

8.1.6 Oblique objects

Oblique objects are arguments that are part of the syntactic valency of the verb (that is, NPs

that express semantic roles specified in the lexical meaning of the verb stem) but which are not

subjects or objects. In Lushootseed, these are introduced by a preposition, in most cases the

“empty” preposition ʔə:

425

(559) a. ʔuʔəɬəd ti ʔaciɬtalbix ʔə ti sʔuladxʷ

ʔu–ʔəɬəd ti ʔaciɬtalbix ʔə ti sʔuladxʷ

PFV–eat SPEC people PR SPEC salmon

‘the people ate the salmon’

(Hess 1995: 32)

b. lecup’ayeq ʔe tiʔiɬ sdiʔdexʷiɬ

lecu–p’ayeq ʔe tiʔiɬ sdiʔdexʷiɬ

CONT–hew PR DIST ATTN–hunting.canoe

‘he was carving out a small hunting canoe’

(Hess 1995: 141, line 34)

c. huy pusiləxʷ ʔə tiʔəʔ skʷup

huy pusil=əxʷ ʔə tiʔəʔ skʷup

SCONJ throw=now PR PROX sucker.fish

‘and then he threw the sucker fish’

(Hess 1998: 88, line 290)

d. ʔuq’ʷəlb tsi luƛ’ ʔə ti sʔuladxʷ

ʔu–q’ʷəl–b tsi luƛ’ ʔə ti sʔuladxʷ

PFV–ready.to.eat–CSMD SPEC:FEM elder PR SPEC salmon

‘the old woman roasted herself a salmon’

(Hess 1995: 32)

The sentences in (559) are examples of bivalent intransitive stems — ʔəɬəd ‘eat, feed on’, p’ayəq

‘carve canoe’, pusil ‘throw’, q’ʷəlb ‘cook/roast something for self’ — whose non-subject

argument appears as an oblique object immediately following the subject. The type of structure

seen in (559d) is particularly common as this is the most regular effect on the valency of a verb

root suffixed with the middle-marker -b (Section 2.1.1.3). As in the examples in (559a) and (d),

the oblique object obligatorily follows an overt subject NP or (as in 562b below) first- or second-

person subject-marker.

Because Lushootseed verbs can never have more than one non-oblique NP in a clause, verbs

that are commonly ditransitive in other languages are monotransitive in Lushootseed, the third

argument (the THEME) being realized as an oblique object:

426

(560) a. ɬildəxʷ tiʔəʔ bibədbədaʔs ʔə tiʔəʔ ʔudaw’

ɬild=əxʷ tiʔəʔ bi–bəd–bədaʔ–s ʔə tiʔəʔ ʔudaw’

give.food=now PROX ATTN–DSTR–child–3PO PR PROX tallow

‘he gave his children tallow’

(Hess 1998: 82, line 135)

b. gʷəl ləkʷyic ʔə tiʔiɬ sʔəɬəd

gʷəl ləkʷ–yi–t–s ʔə tiʔiɬ sʔəɬəd

then eat–DAT–ICS–1SG.OBJ PR DIST food

‘then he ate the food on me’

(Hess 1998: 67, line 60)

Note that in (560b) — which has an applied object licensed by the presence of the dative

applicative (Section 2.1.3) — it is the direct object of the root that is realized as the oblique

object of the applicative stem and the applied argument which is the direct object.

As with other syntactic arguments, oblique objects can be simple noun phrases or more

complex referential expressions, such as those shown in (561):

(561) a. kʷədalikʷəxʷ ʔə tiʔəʔ səsq’ʷəlb ʔə tiʔəʔ bəščəb

kʷəd–alikʷ=əxʷ ʔə tiʔəʔ s=ʔəs–q’ʷəl–b ʔə tiʔəʔ bəščəb

take–ACT=now PR PROX NM=STAT–roast–CSMD PR PROX mink

‘he took of what Mink was roasting’

(Hess 1998: 67, line 48)

b. gʷəl ʔəɬədaxʷ əlgʷəʔ ʔə tiʔəʔ cədiɬ tusɬiltəbs

gʷəl ʔəɬəd=axʷ əlgʷəʔ ʔə tiʔəʔ cədiɬ tu=s=ɬil–t–əb=s

then eat=now PL Pr PROX he PAST=NM=give.food–ICS–PASS=3PO

‘then they ate what food had been given to them’

(Hess 1998: 84, line 180)

By the same token, personal pronouns (Section 2.7.1) can also be oblique objects:

(562) a. gʷəl ləq’ʷuʔcut ʔə tiʔəʔ caadiɬ

gʷəl lə=q’ʷuʔ–t–sut ʔə tiʔəʔ caadiɬ

SCONJ PROG=gathered–ICS–REFL Pr PROX they

‘then he joined them’ (lit. ‘he gathers himself to them’)

(Hess 1998: 79, line 41)

b. ʔəscuucəxʷ čəɬ tiʔiɬ čxʷəluʔ dxʷʔal gʷəlapu

ʔəs–cut–c=əxʷ čəɬ tiʔiɬ čxʷəluʔ dxʷʔal gʷəlapu

STAT–speak–ALTV=now 1PL.SUB DIST whale PR you.guys

‘we have told Whale about you guys’

(Hess 2006: 67, line 600)

427

Unlike direct objects, oblique objects in such constructions can not be promoted to subject by

passivization.

Although oblique objects are commonly introduced by the preposition ʔə, not all constituents

introduced by ʔə are oblique objects: some are agentive complements (see Section 8.1.7 below),

while other are not objects at all but are adjunct adverbial expressions:

(563) a. ƛ’ub čəxʷ ʔə ti adsʔux%ʷ, sgʷəlub

ƛ’ub čəxʷ ʔə ti ad=s=ʔux%ʷ sgʷəlub

good 2SG.SUB PR SPEC 2SG.PO=NM=go pheasant

‘you will be all right as you go, Pheasant’

(Hess 1998: 81, line 98)

b. ʔabačiʔb tiʔəʔ sčətxʷəd ʔə tiʔəʔ ɬiʔɬaʔx%

ʔab–ačiʔ–b tiʔəʔ sčətxʷəd ʔə tiʔəʔ ɬiʔɬaʔx%

extend–hand–MD PROX bear PR PROX small.bowl

‘Black Bear extended his hands over a little bowl’

(Hess 1995: 154, line 72)

c. haˑy bədukʷtubəxʷ ʔə tiʔiʔiɬ ʔə tiʔəʔ suhuys

hay bə=dukʷ–txʷ–b=əxʷ ʔə tiʔ–iʔiɬ ʔə tiʔəʔ s=ʔu–huy=s

INTJ ADD=anger–ECS–PASS=now PR PL–DIST PR PROX NM=PFV–be.done=3PO

‘so they got mad at him for what he had done’

(Hess 1998: 89, line 298)

In the first example in (563a), ʔə introduces a sentential nominal acting as an adverbial

expression of temporal extension, while in (b) it introduces a noun expressing a location (or,

more accurately, a spatial extension), and in (c) sentential nominal expressing motive. These are

clearly not arguments of the verb in the sense of being the expressions of event-participants

assigned semantic roles. Adjunct adverbial phrases such as these will be discussed in more detail

in Section 8.2.7.

8.1.7 Agentive complements

Agentive complements are the agent-phrases of verbs in the passive voice (Section 6.2).

Formally, they look like oblique objects introduced by the preposition ʔə, although they have a

428

number of syntactic properties that distinguish them from ordinary obliques. One of these is their

linear ordering with respect to other syntactic arguments in the clause:

(564) a. kʷukʷucutyitəbəxʷ ʔə tiʔəʔ c’ix%c’ix% tiʔiɬ syaʔyaʔs, sčətxʷəd

kʷukʷucut–yi–t–əb=əxʷ ʔə tiʔəʔ c’ix%c’ix tiʔiɬ syaʔyaʔ–s sčətxʷəd

cook–DAT–ICS–PASS=now PR PROX fish.hawk DIST relative–3PO bear

‘his relative, Black Bear, had food prepared for him by Fish-Hawk’

(Hess 1995: 154, line 68)

b. hay gʷəl, təčtəbaxʷ tiʔəʔ sčətxʷəd ʔə tiʔəʔ c’ix%c’ix%

hay gʷəl təč–t–əb=axʷ tiʔəʔ sčətxʷəd ʔə tiʔəʔ c’ix%c’ix%

INTJ SCONJ roll–ICS–PASS=now PROX bear PR PROX fish.hawk

‘so then Black Bear was rolled [over] by Fish-Hawk’

(Hess 1995: 154, line 84)

As seen here, the agentive complement can either precede or follow the subject of the passive

verb. The order shown in (564a) — with the agentive complement preceding the subject — is the

preferred order, although both are attested in texts.

Like oblique objects, agentive complements can be both ordinary NPs, as in (564), complex

referential expressions, as in (565):

(565) ləcaq’atəb ʔə tiʔiɬ ləxʷəbtəb ʔə tiʔiɬ cədiɬ tiʔiɬ

lə=caq’a–t–əb ʔə tiʔiɬ lə=xʷəb–t–əb ʔə tiʔiɬ cədiɬ tiʔiɬ

PRG–speared–ICS–PASS PR DIST PRG–thrown–ICS–PASS PR DIST he DIST

‘they were impaled by what was thrown at them by those ones there’

(Hess 2006: 62, line 475)

However, unlike obliques, agentive complements can not be first- or second-person pronouns

(Hukari 1976; Jelinek & Demers 1983). This seems to be the only vestige of more thorough-

going restrictions on person and number of subjects and objects found in other Salishan

languages (Czaykowska-Higgins & Kinkade 1998: 39). Agentive complements are also not

legitimate targets for relativization, nor may they be “extracted” from s=nominalizations (i.e.,

structures such as *tiʔiɬ sləxʷəbtəbs tiʔiɬ ‘those who they were throw at by’, based on 565

above, are ungrammatical for the intended gloss): instead, in situations that might require such

David Beck 10-2-7 2:19 PM

Comment: the AgCo second order =

introduction of a secondary topic?

429

constructions, the active voice of the embedded verb is used (cf., tiʔiɬ sləxʷəbəd tiʔiɬ ‘those who

threw at them’). Once again, this restriction differentiates them from ordinary oblique objects.

8.2 Verbal predicates

8.2.1 Monovalent intransitive clauses

very complex subject

(566) gʷəl ləq’axʷ tiʔəʔ tuʔalʔal tudəxʷʔas gʷəʔalʔaləs, gʷəstabəs kʷi cədiɬ tudəxʷʔas əlgʷəʔ

gʷəl lə=q’axʷ tiʔəʔ tu=ʔalʔal tu=dəxʷ=ʔa=s gʷə=ʔalʔal=əs

conj PROG=frozen PROX PAST=house PAST=NM–exist=3PO SBJ=house=3SBJ

gʷə=stab=əs kʷi cədiɬ tu=dəxʷ=ʔa=s əlgʷəʔ

SBJ=what=3SBJ REM he PAST=NM=exist=3PO PL

‘the house where he was, if it were a house, whatever it was, where they had been, froze’

(Hess 1998: 101, line 279)

8.2.2 Transitive clauses

(567) ʔugʷəč’əd ti sqʷəbayʔ

ʔu–gʷəč’–t Ø ti sqʷəbayʔ

PFV–seek–ICS 3SUB PROX dog

‘he/she/it/they looked for the dog’

*‘the dog looked for him/her/it/them’

(Hess 1995: 10)

Just as it regulates the interpretation of transitive clauses with both zero subject and object,

subject-Topic alignment also plays a role in the interpretation of transitive clauses with a single

overt argument, like that in (568c):

(568) a. hiˑkʷ ʔaciɬtalbixʷ tsiʔəʔ sxʷiʔuq’ʷ

hikʷ ʔaciɬtalbixʷ tsiʔəʔ sxʷiʔuq’ʷ

big person PROX:FEM Basket.Ogress

’Basket Ogress

was a big person’

b. ʔuʔux%ʷəxʷ liɬʔal tiʔəʔ lək’awk’aw ʔə tiʔəʔ bək’ʷ stab

ʔu–ʔux%ʷ Ø

liɬ–ʔal tiʔəʔ lə=k’aw–k’aw ʔə tiʔəʔ bək’ʷ stab

PFV–go 3SUB PRLV–at PROX PROG=DSTR–bump PR PROX all what

‘she

is going along (through the brush) bumping into everything’

430

c. ɬčiˑs tiʔəʔ wiw’su

ɬčil–s–Ø Ø

tiʔəʔ wiw’su

arrive–ALTV–3OBJ 3SUB PROX children

‘she

came upon the children’

[AJ Basket Ogress, lines 39 – 41]

As we saw in (722), this episode begins with an intransitive sentence with an overt NP subject,

the referent of this NP becoming the Topic for the next few lines of discourse and the antecedent

of the zero subjects in (568b) and (c). In the latter case, the Topicality of Basket Ogress and the

alignment of Topic and subject guarantees the reading of the sentence as ‘Basket Ogress came

upon the children’, rather than ‘the children came upon Basket Ogress’. This interpretative

principle has become grammaticized at the clausal level so that, even out of context, all transitive

sentences with a single overt argument, that argument is obligatorily interpreted as the direct

object .

8.2.3 Passive and bivalent intransitive clauses

8.2.4 Clauses with multiple objects

Because Lushootseed verbs can never have more than one non-oblique NP in a clause, verbs

that are commonly ditransitive in other languages are monotransitive in Lushootseed, the third

argument being realized as an oblique object:

(569) a. ʔuʔabyid čəd ti č’ač’as ʔə ti sqʷəbayʔ

ʔu–ʔab–yi–d čəd ti č’ač’as ʔə ti sqʷəbayʔ

PFV–extend–DAT–ICS 1SG.SUB SPEC child PR SPEC dog

‘I gave the dog to the boy’

(Hess 1995: 36, ex. 12a)

b. ʔuʔabyitəb čəd ʔə ti č’ač’as ʔə ti sqʷəbayʔ

ʔu–ʔab–yi–t–əb čəd ʔə ti č’ač’as ʔə ti sqʷəbayʔ

PFV–extend–DAT–ICS–PASS 1SG.SUB PR SPEC child PR SPEC dog

‘I was given the dog by the boy’

(Hess 1995: 36, ex. 12c)

The direct object in these expressions is the RECIPIENT rather than the THEME, which surfaces as

an oblique object, making Lushootseed a primary object language in the terms of Dryer (1986).

431

Although it is possible to realize all three arguments in a single clause using a passive

expression as in (569c), this is only the case when the subject is represented by a person-marker.

A non-oblique NP in such constructions is ungrammatical:

(570) a. ʔuʔabyitəb ʔə ti luƛ’ ʔə ti sqʷəbayʔ

ʔu–ʔab–yi–t–əb ʔə ti luƛ’ ʔə ti sqʷəbayʔ

PFV–extend–DAT–ICS–PASS PR SPEC elder PR SPEC dog

‘s/he was given the dog by the old man’

(Hess 1995: 35, ex. 9)

b. ʔuʔabyitəb ʔə ti luƛ’ ti č’ač’as

ʔu–ʔab–yi–t–əb ʔə ti luƛ’ ti č’ač’as

PFV–extend–DAT–ICS–PASS PR SPEC elder SPEC child

‘the child was given it by the old man’

c. ʔuʔabyitəb ti č’ač’as ʔə ti sqʷəbayʔ

ʔu–ʔab–yi–t–əb ti č’ač’as ʔə ti sqʷəbayʔ

PFV–extend–DAT–ICS–PASS SPEC child PR SPEC dog

‘the child was given the dog’

(Hess 1995: 35, ex. 10 – 11)

d. *ʔuʔabyitəb ʔə ti luƛ’ ti č’ač’as ʔə ti sqʷəbayʔ

Note that the restriction here is not against the realization of a particular grammatical role or

roles, but seems to be a restriction against the absolute number of NP arguments in the clause.

(571) a. ɬut’ilibəxʷ čələp ʔə kʷi ɬu[s]əcuucləp

ɬu=t’ilib=əxʷ čələp ʔə kʷi ɬu=s=lə=cut–c=ləp

irr=sing=now 2PL.SUB PR REM IRR=NM=PROG=say–ALTV=2PL.PO

‘you will sing with [the words] you will say to him’

(Hess 1998: 99, line 228)

b. huy č’axʷadəxʷ əlgʷəʔ ʔə tiʔiɬ x%ʷubt

huy č’axʷa–d=əxʷ əlgʷəʔ ʔə tiʔiɬ x%ʷubt

SCONJ club–ICS=now PL PR DIST paddle

‘then he clubbed them with that paddle’

(Hess 2006: 62, line 486)

d. kʷaxʷatubuɬ ʔə tiʔəʔ diʔəʔ

kʷaxʷa–t–ubuɬ ʔə tiʔəʔ diʔəʔ

help–ICS–1PL.OBJ PR PROX here

‘help us with this!’

[AW Basket Ogress, line 114]

432

8.2.5 Clauses with nominal predicate complements

Certain verbs in Lushootseed take nominal predicate complements rather than (or in addition

to) a full NP argument. Nominal predicate complements immediately follow the verb and

precede other NP constituents, as shown in (572):

(572) a. ʔuhuyiləxʷ bəščəb

ʔu–huy–il=əxʷ bəščəb

PFV–be.done–INCH=now mink

‘he became a mink’

(Hess 1995: 118)

b. hay, diɬ dəxʷhuyuds ʔušəbabdxʷ tsiʔəʔ ʔalš ʔə tiʔəʔ suʔəɬadəps

hay diɬ dəxʷ=huyu–d=s ʔušəbabdxʷ tsiʔəʔ

SCONJ FOC ADNM=be.done–ICS=3PO unfortunate.one PROX:FEM

ʔalš–s ʔə tiʔəʔ s=ʔu–ʔəɬadəp=s

cross.sex.sibling–3PO PR PROX NM=PFV–give.feast=3PO

‘so that is how he made his sister poor when she gave a feast’

(Hess 1998: 63, line 66)

c. yəx%i tashuyutəbəxʷ əlgʷəʔ studəq ʔə tiʔəʔ qaw’qs

yəx%i tu=ʔas–huyu–t–əb=əxʷ əlgʷəʔ studəq ʔə tiʔəʔ qaw’qs

because PAST=STAt–be.done–ICS–PASS=now PL slave PR PROX raven

‘because they had been made slaves of Raven’

[HM Star Child, line 96]

Nominal predicate complements such as bəščəb ‘mink’ in (572a) or ʔušəbabdxʷ ‘unfortunate

person’ in (572b) are not introduced by determiners and may not be modified or quantified,

although they can be possessed (572c). They immediately follow the verb and precede any other

objects or adjuncts in the clause (such as tsiʔəʔ ʔalš ‘his sister’, the direct object of the verb in

572b). They also precede overt NP subjects (573a), but follow post-verbal subject-markers and

other S2 particles when the verb is in first position in the sentence, as shown in (573b), and they

remain in place when the particle appears in pre-verbal position (573c):

(573) a. tiləb ʔuhuyil p’q’ac tiʔiɬ sdᶻubuds

tiləb ʔu–huy–il p’q’ac tiʔiɬ s=dᶻubu–d=s

right.away PFV–be.done–INCH rotten.wood DIST NM=kicked–ICS=3PO

‘right away what she kicked became rotten wood’

[DS Star Child, line 142]

433

b. ʔuhuyil čəd kiaʔ ʔal ti dukʷəɬdat

ʔu–huy–il čəd kiaʔ ʔal ti dukʷəɬdat

PFV–be.done–INCH 1SG.SUB grandmother at SPEC yesterday

‘I became a grandmother yesterday’

(Hess 1995: 118)

c. ƛ’ub čəɬ ɬuhuyucutəxʷ ɬukʷaɬ

ƛ’ub čəɬ ɬu=huyu–t–sut=əxʷ ɬukʷaɬ

well 1PL.SUB IRR=be.done–ICS–REFL=now sun

‘we should make ourselves into suns’

[HM Star Child, line 159]

There are, however, a few examples where a predicate complement precedes its verb:

(574) dᶻixʷ ʔaciɬtalbixʷ tuhuyutəb

dᶻixʷ ʔaciɬtalbixʷ tu=ʔu–huyu–t–əb

first person PAST=be.done–ICS–PASS

‘at first [the animals] were made to be like people’

[ML Mink and Tutyika I, line 260]

The environment for this fronting seems to be one of contrastive focus, although more

contextualized examples would be needed to confirm this observation. There are also one or two

examples of clefts formed from predicate nominals:

(575) slələlwaʔs te səshuy ʔə tə sbadil

slə=ləlwaʔs tə s=ʔəs–huy ʔə tə sbadil

PL–sleeping.platform NSPEC NM=STAT–be.done PR NSPEC mountain

‘what the mountains are like [is] sleeping platforms’

(Bates, Hess & Hilbert 1994: 112)

Likewise, nominal predicate complements are also eligible targets for relativization:

(576) x%ʷul’əxʷ ʔukʷədxʷ tiʔəʔ diʔəʔ bəčəɬs bədaʔ tul’ʔal tiʔəʔ tuscəcikʷ ʔup’ic’id

x%ʷul’=əxʷ ʔu–kʷəd–dxʷ tiʔəʔ diʔəʔ bə=s=čəɬ=s bədaʔ

only=now PFV–take–DC PROX here ADD=NM=make=3PO child

tul’–ʔal tiʔəʔ tu=scəcikʷ ʔu–p’ic’i–d

CNTRFG–at PROX PAST=diaper PFV–wrung–ICS

‘she just took this child made from the wrung out diaper’

[HM Star Child 69, line 69]

Not unsurprisingly, “extraction” of the nominal predicate complement requires the

nominalization of the subject clause. This further differentiates nominal predicate complements

from direct objects, which do not require nominalization.

David Beck 10-4-15 11:17 AM

Comment: neat example from the Changer

story (line 121)

˙u d\g„i b\ßç\b çx„a l\–huyu–t–

s ≈\¬ ti s˙uk„uk„

intj you mink 2sg.coord

prog=made–ics–1sg.obj seem

spec game

[XElh ti] was added, but I think

incorrectly

434

Nominal predicate complements also differ from direct objects in that they can not be

passivized; they can also co-occur with overt direct object NPs, as in (572b) above, and appear in

clauses with verbs bearing object- and reflexive-markers:

(577) a. ʔuˑ, dəgʷi bəščəb čxʷa ləhuyuc sʔukʷukʷ

ʔu dəgʷi bəščəb čxʷa lə=huyu–t–s sʔukʷukʷ

INTJ you mink 2SG.SUB PROG=be.done–ICS–1SG.OBJ game

‘oh, it’s you, Mink, you are making fun of me’

(Bates, Hess & Hilbert 1994: 112)

152

b. huy huyucutəxʷ qʷiqʷɬayʔulč ʔa

huy huyu–t–sut=əxʷ qʷi–qʷɬayʔ•ulč ʔa

SCONJ be.done–ICS–REFL=now ATTN–wooden.dish•container be.there

‘then he made himself into a little wooden dish right there’

(Hess 2006: 37, line 377)

Nominal predicate complements also differ from NP arguments of a verb in that they are not

normally referential, but instead have a generic reference, expressing a type of thing rather than

identifying a particular individual or object. Consider the pair of examples in (578):

(578) a. ʔuhuyud čəd tiʔəʔ x%ʷubt

ʔu–huyu–d čəd tiʔəʔ x%ʷubt

PFV–be.done–ICS 1SG.SUB PROX paddle

‘I made a paddle’

(Bates, Hess & Hilbert 1994: 112)

b. ʔuhuyud čəd x%ʷubt

ʔu–huyu–d čəd x%ʷubt

PFV–be.done–ICS 1SG.SUB paddle

‘I made it into a paddle’

(Hess 1995: 118)

The first example here is an ordinary transitive sentence with an NP direct object, this object

referring to a particular PATIENT that (actually or potentially) can be mapped onto an object in the

real world. On the other hand, the same noun, x&ʷubt ‘paddle canoe’, in (578b), does not identify

152

Note that this sentence appears in its original context in Hess (1998: 69, line 121) in amended form as ʔuˑ, dəgʷi

bəščəb čxʷa ləhuyuc [x&əl ti] sʔukʷukʷ ‘oh, it’s you, Mink, you are making me seem like a joke’. This is a stylistic

adjustment — the construction as it appears in the Lushootseed Dictionary is what is heard on tape and is repeated in

this form elsewhere in the same text. The construction as amended would be a more formal version of the same

expression.

435

a PATIENT but instead describes the form which has been given to some (in this case, unnamed)

material or object which would itself be the PATIENT and mapped onto an entity in the real world.

Although clauses with nominal predicate complements are not textually infrequent, there are

only a handful of verbs that can take them. Four of these are based on the same radical √huy ‘be

done, be made, be finished’ and include the radical itself, its inchoative form huyil, its internal

causative form huyud, and its dative applicative form huyid ‘make something for someone’:

(579) a. dəgʷi kʷədaʔ kʷi gʷəƛ’ub gʷəhuyəxʷ ɬukʷaɬ

dəgʷi kʷədaʔ kʷi gʷə=ƛ’ub gʷə=huy=əxʷ ɬukʷaɬ

you PTCL REM SBJ=well SBJ=be.done=now sun

‘maybe the one who should be the sun now [is] you’

[HM Star Child, line 172]

b. ʔəshuyiləxʷ studəq

ʔəs–huy–il=əxʷ studəq

PFV–be.done–INCH–now slave

‘they had become slaves’

(Hess 2006: 60, line 446)

c. ləhuyutəb əlgʷəʔ sdukʷ

lə=huyu–t–əb əlgʷəʔ sdukʷ

PROG=be.done–ICS–PASS PL riff.raff

‘they were just made into nobodies [i.e., lower animals]’

[ML Mink and Tutyika I, line 262]

d. xʷɬub bəhikʷ tiʔiɬ bəshuyitəbs əlgʷəʔ stuləkʷ, gʷəstuləkʷəs

xʷɬub bə=hikʷ tiʔiɬ bə=s=huy–yi–t–əb=s əlgʷəʔ stuləkʷ

even ADD=BIG DIST ADD=NM=be.done–DAT–ICS–PASS=3PO PL river

gʷə=stuləkʷ=əs

SBJ=river=3SBJ

‘an even bigger river was made for them, if it was a river’

(Hess 2006: 36, line 354)

As shown by these examples, the meanings of some of these verbs in nominal predicate

constructions differs somewhat from their meanings in ordinary clauses. Thus, huy ‘be done, be

made’ becomes ‘be as something’, huyil ‘be finished’ means ‘become something’, and huyud

‘make something’ becomes ‘make something like something, make something into something’.

The huyid ‘make something for someone’ has the same English gloss in both types of clauses,

436

although when the patient is expressed as a nominal predicate complement it is interpreted as

non-referential or generic. These four verbs, particularly huyil and huyud, account for the bulk

of the predicate-complement constructions in the present corpus.

Another common verb that takes nominal predicate complements is √šəɬ ‘make something’

(čəɬ in Skagit):

(580) a. ʔučəɬ čəd yiq’ʷus

ʔu–čəɬ č əd yiq’ʷus

PFV–make 1SG.SUB cedar.root.basket

‘I made a cedar-root basket’

(Bates, Hess & Hilbert 1994: 63)

b. ʔuhəliʔdub əlgʷəʔ ʔə tiʔəʔ sušəɬs əlgʷəʔ studəq

ʔu–həliʔ–dxʷ–b əlgʷəʔ ʔə tiʔəʔ s–ʔu–šəɬ–s əlgʷəʔ studəq

PFV–alive–DC–PASS PL PR PROX NM=PFV–make=3PO PL slave

‘they were saved by those who they had made slaves’

(Hess 2006: 63, line 519)

Unlike √huy, however, √šəɬ does not take a direct object, but takes only a predicate complement

expressing the object created by the agent; the material used in the creation is expressed as a

prepositional phrase introduced by tul’ʔal ‘from’:

(581) ʔučəɬ čəd x%ʷubt tul’ʔal tə luƛ’ sx%əlaʔs

ʔu–čəɬ čəd x%ʷubt tul’–ʔal tə luƛ’ sx%əlaʔs

PFV–make 1SG.SUB paddle CNTRFG–at NSPEC old board

‘I made a paddle out of an old board’

(Bates, Hess & Hilbert 1994: 63)

As shown in (576) above, the predicate complement of √šəɬ is accessible to relativization and

requires the formation of an s=nominal. The prepositional phrase expressing the material is also

accessible, although speakers vary between the formation of an s=nominal (582a) or a

dəx&=nominal (582b) in such expressions:

(582) a. ʔuhəliʔdub əlgʷəʔ ʔə tiʔəʔ sušəls əlgʷəʔ studəq

ʔu–həliʔ–dxʷ–b əlgʷəʔ ʔə tiʔəʔ s=ʔu–šəl=s əlgʷəʔ studəq

PFV–alive–DC–PASS PL PR PROX NM=PFV–make=3PO PL slave

‘they were given life by those they had made into slaves’

(Hess 2006: 63, line 519)

437

b. stigʷəd tə dəxʷučəɬ šəbəd

stigʷəd tə dəxʷ=ʔu–čəɬ šəbəd

cedar.withes NSPEC ADNM=PFV–make fish.trap

‘what is used to make fish traps is cedar withes’

(Bates, Hess & Hilbert 1994: 63)

It is not known if this is a stylistic difference, a dialectal variant, or simply free variation.

Like √huy, √šəɬ serves as a base for deriving stems that also take predicate complements:

(583) a. ʔučəɬdxʷ čəd puʔtəd

ʔu–čəɬ–dxʷ čəd puʔtəd

PFV–make–DC 1SG.SUB shirt

‘I made him a shirt’

b. ʔəsčəɬtxʷ əlgʷəʔ hud

ʔəs–čəɬ–txʷ əlgʷəʔ hud

STAT–make–ECS PL fire

‘they have a fire going for it’

(Bates, Hess & Hilbert 1994: 64)

In these forms both -dxʷ (Section 2.1.2.3) and -txʷ (2.1.2.2), which normally act as causatives,

serve as applicatives when combined with this radical, adding a BENEFICIARY/direct object rather

than an AGENT/subject.

Very similar in its syntax to √šəɬ is √təxʷ ‘buy something’:

(584) ɬutəxʷ čəd səpləl

ɬu=təxʷ čəd səpləl

IRR=buy 1SG.SUB bread

‘I’m going to buy some bread’

(Bates, Hess & Hilbert 1994: 224)

Stems derived from √təxʷ also take nominal predicate complements:

(585) a. lətəxʷtxʷ čəd ššiqʷ tsə dʔalš

lə=təxʷ–txʷ čəd ššiqʷ tsə d–ʔalš

PROG=buy–ECS 1SG.SUB hat NSPEC:FEM 1SG.PO–cross.sex.sibling

‘I’ve come to buy my sister a hat’

b. lətəxʷtxʷyid čəd ššiqʷ tsə dʔalš

lə=təxʷ–txʷ–yi–d čəd ššiqʷ tsə d–ʔalš

PROG=buy–ECS–DAT–ICS 1SG.SUB hat NSPEC:FEM 1SG.PO–cross.sex.sibling

‘I’ve come to buy a hat for my sister’

(Bates, Hess & Hilbert 1994: 224)

438

As with √šəɬ, the addition of valency-increasing suffixes to this base adds a BENEFICIARY/object

to the valency of the stem.

The same patterns hold for √ƛ’a ‘to somewhere’:

(586) a. ləƛ’a čəd ʔalʔal

lə=ƛ’a čəd ʔalʔal

PROG=go 1SG.SUB house

‘I’m going home’

b. ləƛ’atxʷ čəd x%ačuʔ, ɬut’it’əb

lə=ƛ’a–txʷ čəd x%ačuʔ ɬu–t’it’əb

PROG=go–ECS 1SG.SUB lake IRR–bathe

‘I’m taking her to the lake to bathe’

(Bates, Hess & Hilbert 1994: 150)

It should be noted, however, that the effect of adding the external causative suffix to √ƛ’a is

consistent with the effect of adding this suffix to other motion verbs (see Section 2.1.2.2 above

for discussion).

A fourth verb that shows similar behaviour is √xʷəɬ ‘lack something’, probably historically

derived from the negative adverb xʷiʔ and the incorporative suffix -əɬ:

(587) a. ƛ’asxʷəɬ čəɬ ƛ’udəxʷucilalikʷs

ƛ’u=ʔas–xʷəɬ čəɬ ƛ’u=dəxʷ–ʔu–cil–alikʷ=s

HAB=STAT–lack 1PL.SUB HAB=ADNM–PFV–dished.up–ACT=3PO

‘we lack something to serve food on’

(Hess 1998: 38, line 389)

b. ʔuxʷəɬiləxʷ əlgʷəʔ sʔəɬəd

ʔu–xʷəɬ–il=əxʷ əlgʷəʔ sʔəɬəd

PFV–lack–INCH=now PL food

‘they had run out of food’

[AW Basket Ogress, line 17]

√xʷəɬ is not attested combined with any of the valency-increasing suffixes, but does combine

with the inchoative -il to form the verb stem xʷəɬil ‘run out of something’ shown in (587b).

An interesting feature of these last four verbs — √šəɬ, √təxʷ, √ƛ’a, and √xʷəɬ — is that the

relevant ordering of the nominal predicate complement and a subject-marker or some other S2

439

particle in the same clause is variable, the predicate complement optionally coming immediately

after the verb:

(588) a. ʔušəɬ ʔu xʷaʔxʷaʔəd b. ʔušəɬ xʷaʔxʷaʔəd ʔu

ʔu–šəɬ ʔu xʷaʔxʷaʔəd ʔu–šəɬ xʷaʔxʷaʔəd ʔu

PFV–make INT clam.basket PFV–make clam.basket INT

‘did you make a clam basket?’ ‘did you make a clam basket?’

c. ʔutəxʷ čəd səpləl d. ʔutəxʷ səpləl čəd

ʔu–təxʷ čəd səpləl ʔu–təxʷ səpləl čəd

PFV–buy 1SG.SUB bread PFV–buy bread 1SG.SUB

‘I bought bread’ ‘I bought bread’

e. ʔuƛ’a čəɬ tawd f. ʔuƛ’a tawd čəɬ

ʔu–ƛ’a č əɬ tawd ʔu–ƛ’a tawd čəɬ

PFV–go 1PL.SUB town PFV–go town 1PL.SUB

‘we went to town’ ‘we went to town’

(Hess 1995: 120)

g. ʔəsx%əɬ čəɬ talə h. ʔəsx%əɬ talə čəɬ

ʔəs–x%əɬ čəɬ talə ʔəs–x%əɬ talə čəɬ

STAT–lack 1PL.SUB money STAT–lack money 1PL.SUB

‘we don’t have any money’ ‘we don’t have any money’

(Bates, Hess & Hilbert 1994: 252)

When immediately following these radicals,

153

the integration of the predicate complement to the

verb stem is such that these complements have been considered incorporated elements on a

structural par with incorporated nouns, and strings such as təxʷsəpləl ‘buy bread’ are often

written as single words. In addition, there is at least one case in the corpus where a compound

headed by one of these three bases is used as a stem for subsequent word-formation:

153

Or stems derived there from:

(i) ʔux%əɬil talə čəɬ

ʔu–x%əɬ–il talə č əɬ

PFV–lack–INCH money 1PL.SUB

‘our money is gone’

(Bates, Hess & Hilbert 1994: 252)

440

(589) gʷəl diɬ dəxʷšəɬt’əbiɬədtubs

gʷəl diɬ dəxʷ=šəɬ–t’əbiɬəd–txʷ–b=s

then FOC ADNM=make–rope–ECS–PASS=3PO

‘and then these were used to make into rope’

(Hess 1998: 80, line 81)

In the absence of further examples, however, it is difficult to say whether or not stem for the

form in (589) in a lexicalized expression, the dependent member of the compound — t’əbiɬəd

‘rope’ — more generally has incorporative properties (perhaps being on its way to becoming a

lexical suffix), or whether the integration of the predicate complement to these four radicals

represents true noun incorporation. A similar issue was raised in the discussion of the suffix -əɬ

‘incorporative’ (Section 2.1.4 above).

8.2.6 Clauses with verbal predicate complements

Some the verbs that can take nominal predicate complements — including √huy ‘be done, be

made’ and its derivatives, and √šəɬ/čəɬ ‘make’ — can also take verbal predicate complements.

These take the form of bare verb stems immediately following the main predicate and

immediately preceding any NP arguments:

(590) a. huy, ƛ’uhuy pədičədəxʷ ʔə kʷi stab ɬuxʷiʔəs ləšudub tiʔiɬ

huy ƛ’u–huy pəd•ič–əd=əxʷ ʔə kʷi stab ɬu–xʷiʔ=əs

SCONJ HAB=be.done earth•covering–ICS=now PR REM what IRR–NEG=3SBJ

lə=šuɬ–dxʷ–b tiʔiɬ

NEGP=see–DC–PASS DIST

‘then she would get them covered with dirt so that it would not be seen’

(Hess 2006: 45, line 96)

b. x%ʷul’ ʔubuusaɬil tiʔiɬ shuyuds sp’ic’ids tiʔəʔ ciʔikʷ ʔə tiʔəʔ bədaʔs

x%ʷul’ ʔu–buus•aɬ–il tiʔiɬ s=huyu–d=s s=p’ic’i–d=s

only PFV–four•times–INCH PROX NM=be.done–ICS=3PO NM=wrung–ICS=3PO

tiʔəʔ ciʔikʷ ʔə tiʔəʔ bədaʔ–s

PROX diaper PR PROX offspring–3PO

‘just four times she did that wringing out of the diaper of her son’

[MW Star Child, line 130]

441

c. huyucutəxʷ x%ʷul’ab ʔə tiʔiɬ ƛ’uhihaʔɬ ƛ’uč’ač’as

huyu–t–sut=əxʷ x%ʷul’ab ʔə tiʔiɬ ƛ’u=hi–haʔɬ ƛ’u=č’ač’as

be.done–ICS–REFL=now seem PR DIST HAB=ATTN–good HAB=child

‘he makes himself seem like a good little child’

(Hess 2006: 41, line 473)

(590a) illustrates a transitive stem, pədičəd ‘cover something with dirt’, serving as the predicate

complement to √huy and preceding the adverbial adjunct, ʔə kʷi stab ɬuxʷiʔəs ləšudub tiʔiɬ ‘so

that it would not be seen’. The sentence in (590b) shows the verb p’ic’id ‘wring something out’

acting as the predicate complement of huyud and taking its own direct object, tiʔəʔ ciʔikʷ ‘the

diaper’. Note that both verbs in the predicate complement construction here are nominalized and

the possessive subject is overtly marked on both verbs. In (590c), the predicate complement is

x&ʷul’ab ‘seem like something’. This verb is a bivalent intransitive verb which takes an oblique

object introduced by the preposition ʔə, in this case tiʔiɬ ƛ’uhihaʔɬ ƛ’uč’ač’as ‘the good little

child’. As shown by both (590b) and (c), the valency and government pattern of verbs used as

predicate complements remains the same as when they are used as main predicates.

Verbs which take verbal predicate complements in the active voice also take such

complements in the passive:

(591) hay gʷəl ɬuhuyutəbəxʷ č’it tiʔiɬ dəxʷʔibəš čəɬ

hay gʷəl ɬu–huyu–t–əb=əxʷ č’it tiʔiɬ dəxʷ=ʔibəš čəɬ

SCONJ then IRR–be.done–ICS–PASS=now be.near DIST ADNM=travel 1PL.PO

‘and so where were are traveling to will be made close by’

(Hess 1995: 147, line 10)

Here, the passive form of the verb huyud ‘make something like something’ taking the

monovalent intransitive radical √č’it ‘nearby’ as a complement. This element immediately

follows the main sentence predicate and precedes the subject, the adjunct nominal tiʔiɬ dəxʷʔibəš

čəɬ ‘where we are traveling’.

In the examples in (590) and (591), the main verb and the predicate complement share the

same subject. It is also possible for the direct object of the main verb to be co-referential with the

some other argument of the predicate complement, as in (592):

442

(592) hay, x%ʷul’əxʷ ʔuhuyud saʔ tiʔiɬ tusʔəɬadəp ʔə tsiʔəʔ ʔušəbabdxʷ k’aʔk’aʔ

hay x%ʷul’=əxʷ ʔu–huyu–d saʔ tiʔiɬ tu=sʔəɬadəp ʔə tsiʔəʔ

SCONJ only=now PFV–be.done–ICS bad DIST PAST=feast PR PROX:FEM

ʔušəbabdxʷ k’aʔk’aʔ

humble.person crow

‘so he simply made bad [i.e., ruined] the feast of poor Crow’

(Hess 1998: 64, line 90)

The sentence in (592) shows the transitive verb huyud ‘make something like something’

governing the monovalent radical √saʔ ‘be bad’. The subject of this radical is the NP tusʔəɬadəp

ʔə tsiʔəʔ ʔušəbabdxʷ k’aʔk’aʔ ‘the feast of poor Crow’, which also corresponds to the nominal

predicate complement of the main verb. In this case, the direct object of the main verb follows

the verbal predicate complement and is realized as a non-oblique NP (i.e., a direct complement).

When the verbal predicate complement is transitive, however, the shared argument is realized as

an oblique:

(593) tučəɬəxʷ ʔaladᶻiʔɬəd ʔə tiʔəʔ diʔəʔ ƛ’uluˑƛ’ ƛ’uqʷɬayʔ ƛ’asɬaq’

tu=čəɬ=əxʷ ʔaladᶻ•iʔɬ–əd ʔə tiʔəʔ diʔəʔ ƛ’u=luƛ’ ƛ’u=qʷɬayʔ

PAST=make=now care.for•child–ICS PR PROX here HAB=old HAB=log

ƛ’u=ʔas–ɬaq’

HAB=stat–fallen

‘she made a babysitter out of that old fallen log’

[MW Star Child, line 7]

In this sentence, the verbal predicate complement is the transitive stem ʔaladᶻiʔɬəd ‘babysit

someone’, whose AGENT (the subject of a main clause based on this verb) is tiʔəʔ diʔəʔ ƛ’uluˑƛ’

ƛ’uqʷɬayʔ ƛ’asɬaq’ ‘that old log that has fallen’. The subject of the main verb, however, is ‘she’

(the mother of Star Child in the text), and so rather than being realized as a non-oblique NP (as in

cases of shared subjects like that in 591 above), the shared argument is expressed as an oblique

object introduced by the preposition ʔə.

154

It should be noted here that the number and variety of examples of verbal predicate complements of the type

discussed here are somewhat limited in the present corpus (essentially, to the examples presented here). A good

many more will have to be uncovered and analyzed before the syntax of these constructions is thoroughly

understood.

443

8.2.7 Adverbs, locatives and adjunct phrases

\˙ib\ß b\k’„ dx„çad

l\–˙ib\ß b\k’„ dx„–çad

prg–travel all toward–where

He traveled everywhere Changer 14 (also bek’w stab)

(594) a. ʔuʔukʷukʷ ʔə tə tib, gʷəl ləɬax%il

ʔu–ʔukʷukʷ ʔə tə tib gʷəl lə=ɬax%il

PFV–play PR SPEC strong SCONJ PROG=night

‘s/he played hard and evening came’

(Hess & Hilbert 1976: I, 50)

b. hiqab čəd ʔuʔibəš ʔə tə lil

hiqab čəd ʔu–ʔibəš ʔə tə lil

excessively 1SG.SUB PFV–walk PR SPEC far

‘I walked too far’

(Hess & Hilbert 1976: I, 51)

note word order:

(595) a. gʷəl lild tul’ʔal tiʔəʔ skəkiʔ tiʔəʔ diʔəʔ č’ač’as

gʷəl lil–d tul’–ʔal tiʔəʔ skəkiʔ tiʔəʔ diʔəʔ č’ač’as

SCONJ be.far–ICS CNTRFGT–at PROX cradleboard PROX here child

‘they remove the boy from the cradle board’

[HM Star Child, line 39]

8.3 Non-verbal predicates

8.3.1 Nominal and nominalized predicates

(596) a. sʔumən’iʔ čəd

155

sʔubədiʔ čəd

hunter 1SG.SUB

‘I am a hunter’

(Hess 1998: 85, line 208)

b. sʔuladxʷ tiʔiɬ

sʔuladxʷ tiʔiɬ

salmon DIST

‘that is a salmon’

(Hess & Hilbert 1976: I, 7)

155

This sentence was pronounced by the storyteller in Raven’s stylized speech, which nasalizes voiced stops.

444

(597) tiʔəʔ tə č’ƛ’aʔ

tiʔəʔ tə č’ƛ’aʔ

PROX NSPEC stone

‘the stone is this one’

(Hess 1995: 81, ex. 5)

(598) wiw’su tiʔəʔ ʔučalad tiʔəʔ sqʷəbayʔ

wiw’su tiʔəʔ ʔu–čalad tiʔəʔ sqʷəbayʔ

children PROX PFV–chase–ICS PROX dog

‘those who chased the dog are the children’

(Hess 1995: 99)

predicate nominals

(599) a. stabtəb tiʔəʔ stigʷəd

stabtəb tiʔəʔ stigʷəd

NM=do–ICS–PASS PROX cedar.withes

‘cedar withes were made’

ha˙k„ s˙\¬\ds \lg„\˙

ha˙k„ s–˙\¬\d–s \lg„\˙

long.time np–eat–3po plural

‘For a long time they ate.’

P&R 183

ads\sb\çalq ˙u qaw’qs

ad–s–\s–b\çalq ˙u qaw’qs

2sg.po–np–stat–kill.game int raven

‘Is [that] game yours, Raven?”’

P&R line 206

g„a(˙) s\shuyalcs

g„a˙ s–˙\s–huy–alc–s

intj np–stat–be.done–cstr–3po

It is his doing.

Coyote’s son line 109

di¬ Ò’usu˙\¬\ds

di¬ Ò’u=s=u–˙\¬\d=s

foc hab=nm=pfv–eat=3po

That is what he would eat. 2 wives, line 210

non-sentential predicate nominals

[h\]la[˙]b ha˙¬\x„ stal≈\x„

h\la˙b ha˙¬–\x„ s–tal≈–\x„

very good–now np–able–now

[Meanwhile the baby] is really nicely capable now.

445

(600) wiˑʔ, x%ʷul’əxʷ čəd ɬusp’ic’ikʷ

wiʔ x%ʷul’=əxʷ čəd ɬu=sp’ic’ikʷ

declare only=now 1SG.SUB IRR=diaper.child

‘she declares, now I will have diaper child’

[DS Star Child, line 159]

(601) ɬustitčulbixʷ čəxʷ

ɬu=s=titčulbixʷ čəxʷ

IRR=NM=small.animal 2SG.SUB

‘you are the one who will be a small animal’

(Hess 2006: 8, line 136)

(602) dsəsʔabyitəb ʔə tiʔiɬ dsqa tiʔəʔ diʔəʔ cəxʷx%ix%ilix%txʷ tiʔəʔ diʔəʔ stawixʷəʔɬ

d=s=ʔəs–ʔab–yi–t–b ʔə tiʔiɬ d–sqa

1SG.PO=NM=STAT–extend–DAT–ICS-PASS PR DIST 1SG.PO–older.brother

tiʔəʔ diʔəʔ d=dəxʷ=x%i–x%ilix%–txʷ tiʔəʔ diʔəʔ stawixʷəʔɬ

PROX he 1SG.PO=ADNM=ATTN–fight–ECS PROX he children

‘this is what my older brother has given me so I can compete with the children’

[MW Star Child, line 100]

y\≈i di¬ ti˙\˙ s\skik\wiçs ti˙i¬ d\x„\s˙uß\bit\bs

y\≈i di¬ ti˙\˙ s–˙\s–ki–k\wiç–s ti˙i¬ d\x„–˙\s–˙uß\b–bi–t–\b–s

because idn det np–stat–attn–hunchback–3po det np2–stat–pity–ss–

ics–pass–3po

because he was a little huchback, she had taken pity on him

Basket ogress, line 27

Nouns bearing possessive markers can be the heads of both referential and predicative

expressions with the exception of nouns with first-person plural possessors, which can not head

predicative expressions:

David Beck 10-2-7 2:19 PM

Comment: might be np, there’s no

possessive marker

446

(603) a. adstaləɬ čəd b. dstaləɬ čəxʷ

ad-staləɬ čəd d–staləɬ čəxʷ

2SG.PO–nephew/niece 1SG.SUB 1SG.PO–nephew/niece 2SG.SUB

‘I am your nephew’ ‘you are my nephew’

c. staləɬs čəd d. staləɬs čəxʷ

staləɬ–s čəd staləɬ–s čəxʷ

nephew/niece–3PO 1SG.SUB nephew/niece–3PO 2SG.SUB

‘I am his/her nephew’ ‘you are his/her nephew’

e. staləɬləp čəd f. *staləɬ čəɬ čəxʷ

staləɬ–ləp čəd staləɬ čəɬ čəxʷ

nephew/niece–2PL.PO 1SG.SUB nephew/niece 1PL.PO 2SG.SUB

‘I am your

nephew’ *‘you are our nephew’

(Hess 1995: 60)

a very complex predicate indeed:

(604) ƛ’ub ʔa ʔal kʷi čad ɬucəxʷəsbeč, ɬudəxʷʔatubšləp, tiʔəʔ sʔəɬəd

ƛ’ub ʔa ʔal kʷi čad ɬu=d=dəxʷ=ʔəs–beč

well exist at REM where IRR=1SG.PO=NM=STAT–fall

ɬu=dəxʷ=ʔa–txʷ–bš=ləp tiʔəʔ sʔəɬəd

IRR=NM=exist–ECS–1SG.OBJ=2PL.PO PROX food

‘there should be food in the place where I am laid, where you folks will put me’

(Hess 1998: 92, line 31)

8.3.2 Other non-verbal predicates

8.4 Interrogatives

8.4.1 Polar questions

Polar questions are questions that can be answered “yes” or “no.” This type of question is

formed in Lushootseed using the interrogative predicate particle ʔu, as in (605):

(605) a. ƛ’uɬčil ʔu tiʔiɬ adʔalalš

ƛ’u=ɬčil ʔu tiʔiɬ ad–ʔal–alš

HAB=arrive INT DIST 2SG.PO–PL–cross.sex.sibling

‘have your brothers arrived?’

(Hess 2006: 45, line 88)

447

b. ʔuˑ ƛ’uc’qʷib ʔu tiʔiɬ sʔušəbabdxʷ sgʷəlub

ʔu ƛ’u=c’qʷib ʔu tiʔiɬ sʔušəbabdxʷ sgʷəlub

INTJ HAB=get.in.on INT DIST poor.guy pheasant

‘oh, is poor Pheasant ever able to share in it?’

(Hess 2006: 45, line 84)

Like other predicate particles (Section 2.5.3), the interrogative appears immediately following a

sentence-initial sentence predicate; with respect to other predicate particles, the interrogative

always follows subject clitics, as in (606a), but generally (although not obligatorily — see 609b

below) precedes most others (606b) with the exception of the particle uʔxʷ (606c), which is

always the first in any string of predicate particles:

(606) a. luud čəxʷ ʔu tiʔiɬ suƛ’əladiʔs

lu–d čəxʷ ʔu tiʔiɬ s=ʔu–ƛ’əladiʔ=s

hear–ICS 2SG.SUB INT DIST NM=PFV–make.noise=3PO

‘do you hear that sound?’

(Hess 2006: 17, line 140)

b. tuɬčil$xʷ ʔu sixʷ

tu=ɬčil=$xʷ ʔu sixʷ

HAB=arrive=NOW INT PTCL

‘have they arrived again?’

(Hess 2006: 44, line 58)

c. ʔəsx%əɬ uʔxʷ čəxʷ ʔu

ʔəs–x%əɬ uʔxʷ čəxʷ ʔu

STAT–sick PTCL 2SG.SUB INT

‘are you still sick?’

(Hess 1995: 88, ex. 7)

As with any predicate particle, ʔu is fronted to follow sentence-initial adverbial elements:

(607) a. ƛ’ub ʔu tabəd

ƛ’ub ʔu tab–d

well INT do–ICS

‘should it be done?’

(Hess 2006: 19, line 161)

b. hikʷ čəd ʔu ƛ’ux%əɬqid

hikʷ čəd ʔu ƛ’u=x%əɬ•qid

big 2SG.SUB INT HAB=sick•head

‘do you generally get severe headaches?’

(Hess 1995: 90, ex. 19)

448

c. ƛ’ub čəɬ ʔu x%ʷul’ ɬut’uk’ʷ

ƛ’ub čəɬ ʔu x%ʷul’ ɬu=t’uk’ʷ

well 1PL.SUB INT only IRR=go.home

‘should we just go home?’

(Hess 1995: 90, ex. 22)

Beyond the presence of the interrogative particle, polar questions in Lushootseed are

syntactically identical to the corresponding affirmative sentence.

The majority of polar questions are in the indicative mood. However, there are one or two

examples of questions in the subjunctive mood:

(608) gʷəyəyəhubəxʷ čəd ʔu

gʷə=yəyəhub=əxʷ čəd ʔu

SBJ=tell.story=now 1SG.SUB INT

‘should I tell a traditional story now?’

(Hess 2006: 10, line 1)

In these cases, the choice of the subjunctive mood seems to indicate that the speaker is seeking

permission or approval, rather than seeking confirmation that an event has occurred or is likely to

occur in the future (cf. the future interrogative in the irrealis mood in 607c).

Polar questions can also be formed on non-verbal predicates such as numerals (609a),

personal pronouns (609b), and s-nominals (609c):

(609) a. buus ʔə sq’aʔšəd kʷi ƛ’uc’əxʷ dxʷʔal kʷi ɬusqits əlgʷəʔ ʔə tiʔəʔ swatixʷəd

buus ʔə sq’aʔšəd kʷi ƛ’uc’=əxʷ dxʷ–ʔal kʷi

four PR moccasin REM tied.in.bunch=now CNTRPT–at REM

ɬu=s=qit=s əlgʷəʔ ʔə tiʔəʔ swatixʷəd

IRR=NM=encircle=3PO PL PR PROX land

‘are there four moccasins that are bundled together so they will encircle the land?’

[DS Star Child, line 351]

b. day’ dᶻəɬ ʔu ʔəca kʷi ʔəsɬuq’ʷɬuq’ʷəyul’qid

day’ dᶻəɬ ʔu ʔəca kʷi ʔəs–ɬuq’ʷ-ɬuq’ʷ•əy•ul’•qid

uniquely PTCL INT I REM STAT–DSTR–peeled•CNN•CNN•head

‘does it seem that I am the only one whose scalp has been peeled?’

156

[ML Mink and Tutyika II, line 89]

156

The character is asking if he is the only one who is bald, he and his cousin having been swallowed by a whale

and having lost their hair as a result.

449

c. adsəsbəčalq ʔu qaw’qs

ad=s=ʔəs–bəč•alq ʔu qaw’qs

2SG.PO=NM=STAT–fall•game INT raven

‘is that your game, Raven?’

(lit. ‘is that the game that you have brought down, Raven?’)

(Hess 1998: 85, line 205)

Note that in (609b) the presence of the adverb day’ ‘uniquely’ causes the fronting of both the

interrogative particle and the particle dᶻəɬ.

One common form of polar question found throughout the texts is a question formed on the

focalizing adverb diɬ, asking for the confirmation of the identity of something:

(610) a. diɬ ʔu tiʔəʔ haʔɬ

diɬ ʔu tiʔəʔ haʔɬ

FOC INT PROX good

‘is the good one this?

c. diɬ ʔu tiʔəʔ diʔəʔ xʷiʔ gʷəsəsʔiq’ʷs

diɬ ʔu tiʔəʔ diʔəʔ xʷiʔ gʷə=s=ʔəs–ʔiq’ʷ=s

FOC INT PROX here NEG SBJ=NM=STAT–swept=3PO

‘is it this one that isn’t cleared?’

[AW Basket Ogress, lines 30–31]

In these constructions, the presupposition of the question is expressed as the syntactic subject,

and the Rhematic element which would correspond to the answer to the question is left unnamed

but is inferable from context. Less commonly, polar questions built on diɬ have an explicit

Rheme and an elided subject:

(611) diɬ ʔu šac’s

diɬ ʔu s=šac’=s

FOC INT NM=end=3PO

‘is that the end?’

[ML Basket Ogress, line 238]

In these cases, rather than asking for confirmation that a particular entity corresponds to some

presupposed or topical identification, the speaker is seeking confirmation of their identification

of a presupposed or topical element. These two patterns parallel exactly the non-interrogative

uses of diɬ, which are discussed in more detail in Section 11.2.1.

450

8.4.2 Information questions

Information questions request specific information, as opposed to seeking an affirmative or

negative answer. In Lushootseed, such questions take the form of sentences whose subject is the

known or Given part of the utterance and whose main predicate is one of the interrogative words

listed in Table 67 in Section 2.6 above. The simplest type of information question asks for the

identity of someone or something, or requests information about its state:

(612) a. gʷat tiʔiɬ stubš

gʷat tiʔiɬ stubš

who DIST man

‘who is that man?’

b. gʷat čəxʷ

gʷat čəxʷ

who 2SG.SUB

‘who are you?’

(Bates, Hess & Hilbert 1994: 97)

c. stab əw’ə tiʔiɬ titčulbixʷ

stab əw’ə tiʔiɬ titčulbixʷ

what PTCL DIST small.animal

’what is that small animal?’

(Bates, Hess & Hilbert 1994: 216)

d. gʷəl ʔəsčal tiʔəʔ sčətxʷəd

gʷəl ʔəs–čal tiʔəʔ sčətxʷəd

SCONJ STAT–how PROX bear

‘and how is Black Bear?’

(Hess 1995: 144, line 35)

The first constituent in these clauses — the interrogative word — is the predicate, and the

predicate is immediately followed by its subject, a simple NP. The mirative predicate particle

əw’ə (Section 2.5.1) seen in (612c) expresses mild surprise and is frequently found associated

with questions.

In addition to being modifiable by predicate particles, interrogative words can also be

modified by adjunct phrases like those in (613):

451

(613) a. gʷatəxʷ ʔə dibəɬ kʷi ɬuʔux%ʷtxʷ kʷi sləx%il

gʷat=əxʷ ʔə dibəɬ kʷi ɬu=ʔux%ʷ–txʷ kʷi sləx%il

who=now PR we REM IRR=go–ECS REM day

‘which [lit. ‘who’] of us will take the daylight?’

(Hilbert & Hess 1977: 18)

b. čad ʔə tiʔəʔ cədiɬ x%əpx%payʔac kʷi sʔas

čad ʔə tiʔəʔ cədiɬ x%əp–x%payʔ•ac kʷi s=ʔa=s

where PR PROX he DSTR–cedar•tree REM NM=be.there=3PO

‘where in these cedars is she?’ [lit. ‘her location [is] where in these cedars?’]

(Hess 2006: 18, line 150)

c. ʔuʔəx%idəxʷ ʔal tuˑˑdiʔ šəq ʔə tə ʔəšab tiʔəʔ sucucuts

ʔu–ʔəx%id=əxʷ ʔal tudiʔ šəq ʔə tə ʔəs–šab tiʔəʔ

PFV–what.happen=now PR DIST.DMA high PR INDEF STAT–dry PROX

s=ʔu–cut–cut=s

NM=PFV–DSTR–say=3PO

‘what is she doing talking way over there up a dry [cedar tree]?’

(Hess 2006: 17, line 145)

In the first two sentences, the scope of the question being asked is restricted using a prepositional

phrase introduced by the general preposition ʔə. In the last, the question is made more specific

with a prepositional phrase specifying a location. Note that in the examples in (613), the subject

phrases are complex nominal expressions — specifically, a headless relative clause in (613a) and

a nominalization in (613b) and (c).

Because the interrogative itself is the sentence predicate and the Given information is

relegated to the subject phrase, the structures in (613b) are quite a common type of construction:

the predicate of the sentence is an interrogative word whose target (the thing being asked about)

is the referent of the complex nominal expression in the subject phrase. The syntax of the subject

phrase follows the same pattern described for relative clauses and nominalizations in Section 7.3.

If the referent of the subject phrase is the subject or object of the verb in the embedded clause,

the subject phrase takes the form of a headless relative clause, as in (614):

(614) a. gʷat kʷi ʔuʔəy’dxʷ ti sqʷəbayʔ

gʷat kʷi ʔu–ʔəy’dxʷ ti sqʷəbayʔ

who REM PFV–find SPEC dog

‘who found the dog?’ (lit. ‘the one who found the dog [is] who?’)

452

b. gʷat kʷi ʔuʔəɬtxʷ čələp

gʷat kʷi ʔu–ʔəɬ–txʷ čələp

who REM PFV–eaten–ECS 2PL.SUB

‘who did you guys feed?’ (lit. ‘the one who you guys fed [is] who?’)

(Hess 1995: 100)

c. gʷat kʷi ʔuʔəy’dub ʔə ti sqʷəbayʔ

gʷat kʷi ʔu–ʔəy’dxʷ–b ʔə ti sqʷəbayʔ

who REM PFV–find–PASS PR SPEC dog

‘who did the dog find?’ (lit. ‘the one found by the dog [is] who?’)

(Hess 1995: 99)

The question in (614a) asks for the identity of the subject of the clause embedded in the subject

phrase, a headless subject-centred relative clause, while (614b) asks for the identity of the direct

object of the object-centred headless relative clause kʷi ʔuʔəɬtxʷ čələp ‘the one that you guys

fed’. Because the subject-phrase in questions of this type is a relative clause, it is subject to the

constraint against the relativization of direct objects in clauses where both subject and object are

third persons (Section 7.4.1); in these situations, the subject phrase must be passivized, as in

(614c). Note that the same constraint prevents the potential ambiguity of (614a), which can not

be interpreted as inquiring after the object of the subject-phrase (i.e., *‘who did the dog find?’).

In questions asking for the identity of an oblique argument, the subject phrase is nominalized

by the proclitic s= (Section 7.4.2.1):

(615) a. gʷatəxʷ kʷi ɬudsq’ʷuʔ ʔal tiʔəʔ

gʷat=əxʷ kʷi ɬu=d=s=q’ʷuʔ ʔal tiʔəʔ

who=now REM IRR=1SG.PO=NM=gathered at PROX

‘who will I get together with in this [place]?’

(lit. ‘the one I’ll get together with here [is] who?’)

[DS Star Child, line 74]

b. stab kʷi gʷəsuʔəɬəds

stab kʷi gʷə=s=ʔu–ʔəɬəd=s

what REM SBJ=NM=PFV–feed.on=3PO

‘what would he eat?’ (lit. ‘the one he would feed on [is] who?’)

(Hess 1995: 143, line 12)

453

c. ʔəstab kʷi gʷədsq’p’ucid

ʔəs–stab kʷi gʷə=d=s=q’p’u–t–sid

STAT–what REM SBJ=1SG.PO=NM=pay–ICS–2SG.OBJ

‘what should I pay you?’ (lit. ‘what I should pay you with [is] what?)

(Hess 2006: 30, line 190)

Nominalized subject phrases are also found with questions targeting predicate-complements of

verbs like huy ‘be done, be made, be finished’ (Sections 8.2.5 and 8.2.6):

(616) a. stab kʷi gʷəshuy čəɬ

stab kʷi gʷə=s=huy čəɬ

what REM SBJ=NM=be.done 1PL.PO

‘what can we do?’

(Hess 2006: 18, line 156)

b. ɬasčaləxʷ kʷi ɬushuy čəɬ

ɬu=ʔas–čal=əxʷ kʷi ɬu=s=huy čəɬ

IRR=STAT–how=now REM IRR=NM=be.done 1PL.PO

‘what are we going to do?’

(Hess 2006: 53, line 279)

c. ʔəsʔəx%id, ʔu dsuq’ʷsuq’ʷaʔ, kʷi gʷədshuy ʔal ti

ʔəs–ʔəx%id ʔu d–suq’ʷ–suq’ʷaʔ kʷi gʷə=d=s=huy

STAT–what.happen INT 1SG.PO–DSTR–sibling REM SBJ=1SG.PO=NM=be.done

ʔal ti

at SPEC

‘what, my little brothers, can I do about this?’

(Hess 2006: 23, line 25)

Unlike questions targeting oblique objects such as those in (615), which are always formed on

stab ‘what?’ or gʷat ‘who?’, questions targeting predicate complements can take either stab, as

in (338a), čal ‘how?’ (338b), or ʔəx&id (338c). It is unclear at present what, if any, difference is

made by the choice of interrogative word in such sentences.

Questions asking for circumstantial information such as motives, etc., that are normally

expressed as adjuncts (Section 8.2.7) are also nominalized. In these cases, either of the two

nominalizing clitics, s= (Section 7.4.2.1) or dəxʷ= (7.4.2.2) may be used, depending on the type

of adjunct being questioned. Questions based on the interrogatives čad ‘where?’ (that is, čad

454

‘where’, dxʷčad ‘to where?’, tul’čad ‘from where?’, and liɬčad ‘which way?’), almost

invariably take s=nominals as their subjects:

157

(617) a. huy gʷəl čad kʷi sʔəy’dubs ʔə tiʔəʔ cədiɬ

huy gʷəl čad kʷi s=ʔəy’dxʷ–b=s ʔə tiʔəʔ cədiɬ

SCONJ SCONJ where REM NM=find–DC=3PO PR PROX he

‘so then, where were they found by this one?’

(Hess 1998: 74, line 224)

b. liɬčad kʷi ɬadsuʔux%ʷ

liɬ–čad kʷi ɬu=ad=s=ʔu–ʔux%ʷ

PRLV–where REM IRR=2SG.PO=NM=PFV–go

‘which way will you go?’

(Hess 1995: 105)

c. tul’–čad kʷi skʷədxʷs əlgʷəʔ

tul’–čad kʷi s=kʷəd–dxʷ=s əlgʷəʔ

CNTRFG–where REM NM=taken–DC=3PO PL

‘from where did they manage to get it?’

(Hess 1998: 83, line 162)

Questions relating to temporal adjuncts also generally take s=nominals as subjects, as in (618):

(618) a. pədtab kʷi ɬudšudubicid

pədtab kʷi ɬu=d=s=šuɬ–dxʷ–bicid

when REM IRR=1SG.PO=NM=see–DC–2SG.OBJ

‘when will I see you?’

(Bates, Hess & Hilbert 1994: 216)

157

There is one exception to this generalization in the present corpus:

(i) čad swatixʷəd tiʔəʔ dəxʷʔ atubs

čad swatixʷəd tiʔəʔ dəxʷ=ʔa–txʷ–b=s

where land PROX ADNM=be.there–ECS–PASS=3PO

‘where in the world had they been placed?’

(Hess 2006: 54, line 293)

However, the structure of this sentence is unclear, as its predicate appears to be composed of an interrogative word,

čad ‘where?’ and a bare nominal, swatixʷəd ‘land, world’ which may be functioning as a predicate complement of

some kind (see Section 8.2.5). However, there are no other examples of this type available, and any generalizations

that might be drawn here would be little more than speculation.

455

b. x%ʷul’əxʷ ʔəsʔəx%id tiʔiɬ sʔa ʔə cədiɬ tučəgʷas dxʷʔal t’aq’t

x%ʷul’=əxʷ ʔəs–ʔəx%id tiʔiɬ s=ʔa ʔə cədiɬ tu=čəgʷas–s

only=now PFV–what.happen DIST NM=be.there PR he PAST=wife–3PO

dxʷ–ʔal t’aq’t

CNTRPT–at inland

‘just how long was his wife there up from shore?’

(Hess 2006: 16, line 126)

Likewise, questions about the motives for an action usually have this type of subject as well:

(619) a. bələčayɬəxʷ čələp kʷi sʔuʔux%ləp

bə=lə=čayɬ=əxʷ čələp kʷi s=ʔu–ʔux%=ləp

ADD=PROG=go.for.what=now 2PL.SUB REM NM=PFV–go=2PL.PO

‘what business did you folks have in going there?’

(Hess 1998: 93, line 80)

b. ʔəsʔəx%id əw’ə tiʔiɬ ƛ’udsəsqəldub ʔə tiʔiɬ cəxʷʔacəc sɬəɬadəyʔ dxʷʔal kʷi gʷədsʔux%ʷ

dxʷʔistaʔulgʷədxʷ gʷəšayiləd

158

ʔəs–ʔəx%id əw’ə tiʔiɬ ƛ’u=d=s=ʔəs–qəldxʷ–b ʔə tiʔiɬ

STAT–what.happen PTCL DIST HAB=1SG.SUB=NM=STAT–stop–PASS PR DIST

d=dəxʷ=ʔacəc sɬə–ɬadəyʔ dxʷ–ʔal kʷi gʷə=d=s=ʔux%ʷ

1SG.PO=ADNM=be.located PL–woman CNTRPT–at REM SBJ=1SG.SUB=NM=go

dxʷ–ʔistaʔ•ulgʷədxʷ gʷə=šayil=əd

CNTRPT–same•land SBJ=hunt=1SG.SBJ

‘why do these women here stop me from going towards this same land when I hunt?’

[HM Star Child, line 84]

There are also some examples of questions asking about means that have s=nominal subjects:

158

Compare this sentence with the sentence in (i), which takes a subject in dəxʷ= rather than s=:

(i) ʔuˑ, ʔəsʔəx%id əw’ə tiʔiɬ dəxʷx%aʔx%aʔtubs ʔə tiʔiʔiɬ sɬəɬadəyʔ gʷə dsuʔibəš dxʷ ʔaɬx%adulgʷədxʷ ʔə tiʔiɬ

dsuxʷʔxʷiʔ

ʔu ʔəs–ʔəx%id əw’ə tiʔiɬ dəxʷ=x%aʔx%aʔ–txʷ–b=s ʔə tiʔ–iʔiɬ sɬə–ɬadəyʔ

INTJ STAT–what.happen PTCL DIST ADNM=forbidden–ECS–PASS=3PO PR PL–DIST PL–woman

gʷə=d=s=ʔu–ʔibəš dxʷ–ʔaɬx%ad•ulgʷədxʷ ʔə tiʔiɬ d=s=ʔu–xʷ ʔxʷiʔ

SBJ=1SG.PO=NM=PFV–travel CNTRPT–downstream•land PR DIST 1SG.PO=NM=PFV–forage

‘oh, why is my travel in the land downstream forbidden by those women when I hunt?’

[MW Star Child, line 34]

In this case, the use of dəxʷ= instead of s= may indicate that the focus of the question is on the cause of the

women’s prohibition rather than the motive for it, which is the focus in (619). This would make (i) an instance of the

regular use of dəxʷ= in questions about causes, illustrated in (623) below.

456

(620) a. ʔəsčaləxʷ kʷi ɬushuys

ʔəs–čal=əxʷ kʷi ɬu=s=huy=s

STAT–how=now REM IRR=NM=be.done=3PO

‘how will he manage?’

(Hess 1995: 105)

b. gʷəl ʔəsčaləxʷ həw’ə kʷi tushuys əlgʷəʔ

gʷəl ʔəs–čal=əxʷ həw’ə kʷi tu=s=huy=s əlgʷəʔ

SCONJ STAT–how=now PTCL REM PAST=NM=be.done=3PO PL

‘but how did they deal with it?’

(Hess 2006: 41, line 486)

However, questions asking about both time and means are also found with subjects nominalized

by the adjunct nominalizer, dəxʷ=:

(621) a. pədtab kʷi adəxʷəxʷcutəb gʷəsɬcils

pədtab kʷi ad=dəxʷ=ʔəs–dxʷ–cut–əb gʷə=s=ɬcil=s

when REM 2SG.PO=ADNM=STAT–CTD–say–DSD SBJ=NM=arrive=3PO

‘when do you think he will arrive?’

(Bates, Hess & Hilbert 1994: 216)

b. ʔəsčal kʷi dəxʷgʷəlaltəbs

ʔəs–čal kʷi dəxʷ=gʷəlal–t–əb=s

STAT–how REM ADNM=punished–ICS–PASS=3PO

‘how was she punished?’

(Hess 2006: 80, line 924)

c. tasčaləxʷ kʷi tudəxʷʔibəšəxʷ

tu=ʔas–čal=əxʷ kʷi tu=dəxʷ=ʔibəš=s=əxʷ

PAST=STAT–how=now REM PAST=ADNM=travel=3PO=now

‘how did they get there?’

[ML Basket Ogress, line 217]

In the first of these examples, the use of the adjunct nominalizer may follow from the fact that

the interrogative word is questioning an adjunct of a complement of the main verb in the subject

clause, dxʷcutəb ‘think something’, rather than an adjunct of the main verb itself (i.e., the

question asks about the time of arrival, not the time of thinking). The uses of dəxʷ= in (621b)

and (c) are more problematic, however, as they seem exactly to parallel the uses in the questions

in (620). The parallelism may be a result of the translation into English, or it may simply be an

457

example of free variation. The resolution to this question will have to await the discovery of

further sentences of this type with dəxʷ=.

More consistently, the adjunct nominalizer dəxʷ= is used in the subject-phrases of questions

asking about other types of adjuncts, such as instruments:

(622) a. stab tiʔəʔ dəxʷut’uc’utəbs tiʔəʔ qʷiqʷqʷistay’bixʷ

stab tiʔəʔ dəxʷ=ʔu–t’uc’u–t–əb=s tiʔəʔ qʷi–qʷ–qʷistay’bixʷ

what PROX ADNM=PFV–be.shot–ICS–PASS=3PO PROX ATTN–ATTN–dwarf

‘what have the dwarves been shot with?’

(Hess 2006: 76, line 804)

b. stabəxʷ kʷi gʷədəxʷč’axʷalikʷ čəɬ ʔə tiʔəʔ bəkʷ stab, tiʔəʔ stab, tiʔəʔ haʔɬ ʔə ti buʔqʷ

stab=əxʷ kʷi gʷə=dəxʷ=č’axʷ–alikʷ čəɬ ʔə tiʔəʔ bəkʷ stab

what=now REM SBJ=ADNM=clubbed–ACT 1PL.PO PR PROX all what

tiʔəʔ stab tiʔəʔ haʔɬ ʔə ti buʔqʷ

PROX what PROX good PR SPEC waterfowl

‘what can we use to club all these things, these things, these good waterfowl?’

(Hess 2006: 76, line 812)

Similarly, questions about causation are also formed with dəxʷ=nominals as their subjects:

(623) stab tiʔəʔ dəxʷuʔatəbəds əlgʷəʔ

stab tiʔəʔ dəxʷ=ʔu–ʔatəbəd=s əlgʷəʔ

what PROX ADNM=PFV–die=3PO PL

‘what was causing them to die?’

(Hess 2006: 63, line 499)

Because the questions in (622) and (623) ask about the identity of specific things, the

interrogative word used to formulate the question is stab ‘what?’. While this is the most common

case for asking about instruments (which are, after all, objects), causes are more often events, in

which case the question word used is ʔəx&id ‘what happens?’:

(624) a. ʔəsʔəx%id kʷi dəxʷəstagʷəxʷ ʔə tiʔəʔ qaw’qs

ʔəs–ʔəx%id kʷi dəxʷ=ʔəs–tagʷ=əxʷ ʔə tiʔəʔ qaw’qs

STAT–what.happen REM ADNM=STAT–hungry=now PR PROX raven

‘why is Raven hungry?’

(Hess 1995: 105)

458

b. ʔuʔəx%id əw’ə tiʔiɬ adəxʷʔitut

ʔu–ʔəx%id əw’ə tiʔiɬ ad=dəxʷ=ʔitut

PFV–what.happen PTCL DIST 2SG.SPO=ADNM=sleep

‘why did you fall asleep?’

[AW Basket Ogress, line 95]

c. ʔuʔəx%id kʷi dəxʷx%əɬ ʔə tsiʔəʔ x%ʷuʔx%ʷəyʔ

ʔu–ʔəx%id kʷi dəxʷ=x%əɬ ʔə tsiʔəʔ x%ʷuʔx%ʷəyʔ

PFV–what.happen REM ADNM=sick PR PROX:FEM helldiver

‘how did Helldiver get sick?’

(Hess 2006: 11, line 7)

As shown in these examples, questions of this type are typically translated into English as why?

(a word with no exact Lushootseed equivalent) or how?, although neither of these translations is

appropriate for ʔəx&id in isolation or in other contexts. The Lushootseed word that translates most

directly as how?, čal, on the other hand, is much more frequently associated with questions

about means that take s=nominals as their subject, such as those shown in (620) above.

8.5 Imperatives

Like all languages, Lushootseed has a variety of means of issuing commands and directives

to second persons, the most direct of these being the use of the imperative mood, which is

marked by a special set of subject markers (Section 8.1.1).

159

In the second person, the

imperative subject marker is Ø, resulting in an surface form that is homophonous with the third-

person indicative form:

(625) a. lilcut

lil–t–sut Ø

far–ICS–REFL 2SG.IMP

‘go away!’

(Hess 1998: 83, line 152)

b. hay q’ʷaʔq’ʷab

hay q’ʷaʔq’ʷab Ø

SCONJ bark.like.seal 2SG.IMP

‘so now bark like a seal!’

(Hess 2006: 48, line 165)

159

The imperative mood seems to be largely restricted to the affirmative; negative commands take the form of

negative assertions, and are discussed in the Section 8.6.5 under the general heading of negation.

459

c. ʔuhaʔliʔɬ

ʔu–haʔl•iʔɬ Ø

PFV–stop.from.crying•child 2SG.IMP

‘tend to the baby!’

(Hess 2006: 5, line 71)

Imperatives are found only in the imperfective and perfective aspects, the vast majority being in

the imperfective. Not infrequently, imperatives are accompanied by a vocative aimed at the

addressee:

(626) a. lilcut wiw’su

lil–t–sut wiw’su

far–ICS–REFL children

‘go away, children!’

(Hess 1998: 93, line 61)

b. ʔux%alikʷ ʔə kʷi siq’wayuʔ, qiʔqəladiʔ, dəxʷɬalš čəɬ tsiʔəʔ adskʷuy

ʔux%–alikʷ ʔə kʷi siq’wayuʔ qiʔqəladiʔ dəxʷ=ɬal–š čəɬ

go–ACT PR REM forked.stick Qiqeladi ADNM=remove.from.fire–ICS 1PL.PO

tsiʔəʔ ad–skʷuy

PROX:FEM 2SG.PO–mother

‘go for a forked stick, Qiqeladi, so we can get your mother out of the fire!’

[ML Basket Ogress, line 101]

The vocative is generally set off prosodically, as indicated by the commas in (626b). (626b) also

shows that non-subject arguments are realized normally, here as an oblique object in a

prepositional phrase, following the verb but preceding the vocative. Direct objects are also

realized in just as in indicative clauses, as either NPs directly following the verb (627a) or as

object suffixes (627b):

(627) a. ʔəƛ’txʷ kʷi adɬaʔx%

ʔəƛ’–txʷ kʷi ad–ɬaʔx%

come–ECS REM 2SG.PO–platter

‘bring your platter!’

(Hess 2006: 43, line 22)

b. kʷaxʷatubuɬ ʔə tiʔəʔ diʔəʔ

kʷaxʷa–t–ubuɬ ʔə tiʔəʔ diʔəʔ

help–ICS–1PL.OBJ PR PROX here

‘help us with this here (thing)!’

[AW Basket Ogress, line 114]

460

As in the previous examples, these forms are homophonous with indicative clauses with third-

person subjects.

Imperatives directed at second-person plural addressees take the second-person plural

imperative subject maker ɬi:

(628) a. lilcut ɬi

lil–t–sut ɬi

far–ICS–REFL 2PL.IMP

‘go away, you guys!’

(Hess 1998: 93, line 60)

b. šuuc ɬi

šuɬ–c ɬi

see–ALTV 2PL.IMP

‘look at it, you guys!’

c. hiwil ɬi

hiwil ɬi

go.ahead 2PL.IMP

‘go on, you guys!’

(Hess 1998: 83, lines 147–148)

As with singular imperatives, plural imperatives can also have a vocative NP:

(629) a. ʔu, ƛ’əld ɬi, dsuq’ʷsuq’ʷaʔ

ʔu ƛ’əld ɬi d–suq’ʷ–suq’ʷaʔ

INTJ ignore 2PL.IMP 1SG.PO–DSTR–younger.sibling

‘oh, never mind my younger brothers!’

(Hess 2006: 23, line 35)

b. ɬalibš ɬi wiw’su, ɬuʔabaqtubuɬəd čəd

ɬal–š–bš ɬi wiw’su ɬu=ʔabaq–t–ubuɬəd čəd

remove.from.fire–ICS–1SG.OBJ 2PL.IMP children IRR=return–ICS–2PL.OBJ 1SG.SUB

‘get me out of the fire, children, I will take you back (home)!’

[AJ Basket Ogress, line 102]

The non-subject arguments of plural imperatives, as well as their adjuncts, are also realized like

objects in the indicative mood. The example in (629b), for instance, bears an object-marker. The

examples in (630a) and (b) show plural imperatives with object NPs, while the example in (630c)

illustrates a plural imperative with a locative PP:

461

(630) a. ʔuˑ, kʷədad ɬi ti siq’wayuʔ

ʔu kʷədad ɬi ti siq’wayuʔ

INTJ taken–ICS 2PL.IMP SPEC forked.stick

‘oh, take the forked stick, you guys!’

[AJ Basket Ogress, line 104]

b. ɬač’ad ɬi tiʔiɬ hud, stawixʷaʔɬ

ɬač’a–d ɬi tiʔiɬ hud stawixʷaʔɬ

extinguished–ICS 2PL.IMP DIST fire children

‘put out the fire, children!’

[LA Basket Ogress, line 76]

c. kʷaxʷac ɬi dxʷʔal gʷədskiisdubut ʔal tiʔəʔ

kʷaxʷa–t–s ɬi dxʷ–ʔal gʷə=d=s=kiis–dxʷ–but ʔal tiʔəʔ

help–ICS–1SG.OBJ 2PL.IMP CNTRPT–at SBJ=1SG.PO=NM=stand–DC–REFL at PROX

‘help me so that I can stand up in here, you guys!’

[MS Basket Ogress, line 63]

As expected, the subject marker immediately follows the verb and precedes the object. Note also

that the vocative in (630b) follows the direct object.

Occasionally, plural imperatives are found in a series. In these cases, the subject-marker

appears only once, following the first verb:

(631) hiwil ɬi, šuuc

hiwil ɬi šuɬ–c

go.ahead 2PL.SUB see–ALTV

‘go on, you guys, look at it!’

(Hess 1998: 83, line 146)

There are also attestations of sequential imperatives in the singular in the texts, but the fact that

the singular imperative subject marker is zero makes these, barring a study of their prosody,

indistinguishable from two commands contained in separate clauses (which is, in general, how

they are transcribed, on separate line with exclamation marks following the gloss and, in some

cases, the Lushootseed expression).

Aside from the true imperatives, a very frequent strategy for issuing commands to a singular

addressee is through the use of the second-person singular indicative:

462

(632) a. dᶻubalikʷ čəxʷ

dᶻubalikʷ čəxʷ

dance 2SG.SUB

‘you dance!’

[AJ Basket Ogress, line 86]

b. qəliˑltxʷ čəxʷ tsi adčəgʷas

qəl–il–txʷ čəxʷ tsi ad–čəgʷas

bad–INCH–ECS=now 2SG.SUB SPEC:FEM 2SG.PO–wife

‘you make your wife stop!’

(Hess 2006: 5, line 63)

Unlike true imperatives, this type of command can be issued in aspects other than the

imperfective and perfective:

(633) ʔəsxʷəctxʷ čəxʷ tiʔiɬ adbitaʔs, tiʔiɬ adsəsx%əqšad, tiʔiɬ adƛ’əwc’lax%ad

ʔəs–xʷəc–txʷ čəxʷ tiʔiɬ ad–bitaʔs tiʔiɬ

STAT–removed–ECS 2SG.SUB DIST 2SG.PO–breechcloth DIST

ad=s=ʔəs–x%əq•šad tiʔiɬ ad–ƛ’əwc’lax%ad

2SG.PO=NM=STAT–wrapped•leg DIST 2SG.PO–armbands

‘Take off your breechcloth, your leg wraps, your leg wrappings, your armbands.’

(Hess 2006: ?Coyote’s son 79)

It is not entirely clear what conditions govern the choice of mood in commands. It might be

supposed that it has to do with politeness or indirection, although there are examples where an

imperative and a indicative occur together in the same sentence:

(634) a. hiwiləxʷ čxʷa təqad tə šəgʷɬ

hiwil=əxʷ Ø čxʷa təqa–d tə šəgʷɬ

go.ahead=now 2SG.IMP 2SG.COORD closed–ICS NSPEC door

‘go ahead and you close the door!’

b. hiwiləxʷ čəxʷ təqad tə šəgʷɬ

hiwil=əxʷ čəxʷ təqa–d Ø tə šəgʷɬ

go.ahead=now 2SG.SUB closed–ICS 2SG.IMP NSPEC door

‘you go ahead and close the door!’

(Hess & Hilbert 1976: I, 43)

160

In (634a), there are two clauses in a coordinate relation, the first being in the imperative mood

and the second, introduced by the second-person singular coordinative subject-marker in the

160

The (a) and (b) forms of the sentences reflect the possible variation noted in footnote 2. The transcription (čxʷa

instead of čəxʷa) has been revised to follow current practices.

463

indicative mood. (634b), on the other hand, has the first of the two verbs in the indicative mood

and the second, apparently, in the imperative — although it could also simple represent a series

of two verbs in the indicative mood “sharing” the second-person subject clitic (cf. Section 9.1).

Plain indicatives used as commands directed at plural addressees are much less frequent, but

they are attested in the corpus:

(635) a. ʔəsqʷibəxʷ čələp

ʔəs–qʷib=əxʷ čələp

STAT–prepared=now 2PL.SUB

‘you folks be ready!’

(Hess 2006: 66, line 588)

b. hay čələp ɬuʔux%ʷcəxʷ čələpa ɬugʷiidəxʷ

hay čələp ɬu=ʔux%ʷ–c=əxʷ čələpa ɬu=gʷi–d=əxʷ

SCONJ 2PL.SUB IRR=go–ALTV=now 2PL.COORD IRR=invited–ICS=now

‘next you guys will go and you guys will invite them’

(Hess 2006: 73, line 738)

Note that the example in (635b) consists of two clauses, both with imperative force, the first in

the present indicative and the second in the future. This use of the future tense in imperative

contexts is found quite frequently in texts directed at both singular and plural addressees, as are

expressions with the adverb ƛ’ub ‘well’, which in these contexts takes on a modal sense of

‘should’ or ‘ought to’:

(636) a. ƛ’ub čələp ʔušuuc

ƛ’ub čələp ʔu–šuɬ–c

well 2PL.SUB PFV–see–ALTV

‘you guys should look him over!’

(Hess 1998: 98, line 190)

b. ƛ’ubəxʷ čəxʷ ʔuhay’əd tsi adʔalš

ƛ’ub=əxʷ čəxʷ ʔu–hay’əd tsi ad–ʔalš

well=now 2PL.SUB PFV–pay.attention.to SPEC:FEM 2SG.PO–sibling

‘you had better pay attention to your sister’

(Hess 1998: 60, line 11)

This imperative force of ƛ’ub in such expressions, however, is purely pragmatically-determined,

and the same adverb is found in contexts such as (637a), where the speaker is offering advice to

464

the addressee, and (637b), where the speaker is merely offering an opinion as to what ought to

happen in a given situation:

(637) a. ƛ’ub čəxʷ ʔukʷədad

ƛ’ub čəxʷ ʔu–kʷəda–d

well 2SG.SUB PFV–taken–ICS

‘you should take her’

(Hess 1998: 98, line 202)

b. ƛ’ub čəxʷ ʔubiqʷyitəb ʔə tsiʔacəc tubədaʔs

ƛ’ub čəxʷ ʔu–biqʷyit–b ʔə tsiʔacəc tu=bədaʔ–s

well 2SG.SUB PFV–permit–PASS PR UNQ:FEM PAST=offspring–3PO

‘you should be permitted the deceased’s daughter’

(Hess 1998: 98, line 205)

Thus, ƛ’ub in and of itself can not be considered an imperative marker, but instead offers the

speaker an alternative to issuing a direct command in the imperative mood. As with the use of

the second-person indicative, the fact that ƛ’ub constructions with imperative force are textually

more frequent with plural addressees than with singular addressees may well reflect issues of

politeness and social restrictions on the use of the imperative and direct commands.

8.6 Negation

The primary marker of negation in Lushootseed is the negative adverb xʷiʔ, which — like

other lexical adverbs — is attested both as a syntactic predicate and as an adverbial modifier.

When used as a sentence predicate in its own right, it functions to negate the presence of the

referent of its subject at a particular location:

(638) a. xʷiʔəxʷ tsiʔiɬ čəgʷas

xʷiʔ=əxʷ tsiʔiɬ čəgʷas–s

NEG=now DIST:FEM wife–3PO

‘his wife was not there’

(Hess 2006: 8, line 144)

b. bəxʷiʔəxʷ tiʔiɬ ƛ’udsəsq’ʷəladup sʔəɬəd

bə=xʷiʔ=əxʷ tiʔiɬ ƛ’u=d=s=ʔəs–q’ʷəl•adup sʔəɬəd

ADD=NEG=now DIST HAB=1SG.PO=NM=STAT–cooked•land food

‘again the food which I had baked in the ground is gone’

(Hess 2006: 39, line 413)

465

c. xʷiʔəxʷ sixʷ ti tusʔəɬəd čəɬ

xʷiʔ=əxʷ sixʷ ti tu=sʔəɬəd čəɬ

NEG=now PTCL SPEC PAST=food 1PL.PO

‘again our food is gone’

(Hess 2006: 41, line 477)

In these environments, xʷiʔ is glossed along the lines of ‘not be there’, which is consistent with

the meaning xʷiʔ has as the radical of derived forms such as xʷiʔil ‘be gone, become not-there’:

(639) a. bək’ʷ ʔuxʷiʔil

bək’ʷ ʔu–xʷiʔ–il

all PFV–NEG–INCH

‘it’s all gone’

(Bates, Hess & Hilbert 1994: 252)

b. gʷəl x%ɬaxʷ ti ləxʷiʔil tiʔiɬ tusəstabs

gʷəl x%ɬ=axʷ ti lə=xʷiʔ–il tiʔiɬ tu=s=ʔəs–tab=s

SCONJ seemingly=now PROG=NEG–INCH DIST PAST=NM=STAT–do=3PO

‘and it was as though what he had done was all gone’

(Hess 1998: 99, line 215)

The same meaning of the radical is also seen in a few other forms (e.g., xʷiʔalusbid ‘feel the

absence of something’) and with one use of xʷiʔtxʷ ‘cause not to be there’:

161

(640) xʷiʔtubəxʷ

xʷiʔ–txʷ–b=əxʷ

NEG–ECS–PASS=now

‘he is not here now (because of what happened)’

[HM Star Child, line 57]

However, other uses of xʷiʔtxʷ have a slightly different reading:

(641) xʷiʔtubəxʷ ʔə tə pastəd. xʷiʔtxʷəxʷ kʷi tubəkʷ tusʔəɬəd

xʷiʔ–txʷ–b=əxʷ ʔə tə pastəd

NEG–ECS–PASS=now PR NSPEC white.person

xʷiʔ–txʷ=əxʷ kʷi tu=bəkʷ tu=sʔəɬəd

NEG–ECS=now REM PAST=all PAST=food

’the white people have caused it to be no more. There is no more (Indian) food left’

(Bates, Hess & Hilbert 1994: 252)

161

Another use of this verb form in which the radical seems to contribute the more general negative sense is

discussed in Section 8.6.5.

466

Here, xʷiʔtxʷ means ‘cause not to be, cause not to exist’, the contribution of the negative radical

being ‘not be’ or ‘not exist’ rather than ‘not be there’, as does the use of xʷiʔtxʷ in the imperative

mood to form negative commands or admonishments (see Section 8.6.5 below). These facts

seem more in keeping with the uses of xʷiʔ as a general-purpose negator illustrated in the

sections below, and may be an indication that etymologically-speaking xʷiʔ originated as some

kind of negative adverb or locative expression, and underwent a metaphorical extension from

‘not be there’ to ‘not be’ and then finally took on the more general meaning ‘not’ in modern

Lushootseed. With this final, most general meaning, xʷiʔ is found in a wide range of fairly

specialized syntactic constructions, each of which will be discussed in more detail in the sections

that follow.

8.6.1 Adverbial negation

The most straightforward negative construction in Lushootseed from the English perspective

is adverbial negation, in which the negative xʷiʔ is used to negate the proposition expressed by

the syntactic predicate, as in the examples in (642):

(642) a. gʷəl xʷiʔ ɬuləx%əb dxʷʔal dəgʷi

gʷəl xʷiʔ ɬu=lə=x%əb dxʷ–ʔal dəgʷi

SCONJ NEG IRR=NEGP=heavy CNTRPT–at you

‘it will not be heavy for you’

(Hess 1998: 81, line 92)

b. xʷiʔəxʷ ləɬild tsiʔəʔ čəgʷas

xʷiʔ=əxʷ lə=ɬil–d tsiʔəʔ čəgʷas–s

NEG=now NEGP=give.food–ICS PROX:FEM wife–3PO

‘he didn’t give any of the food to his wife’

(Hess 1998: 84, line 168)

Like other adverbs, xʷiʔ appears in these constructions in pre-predicate position and attracts

sentence-second clitics and predicate particles (Section 2.5.3) such as the matrix subject markers

seen in (643a) and (b):

467

(643) a. xʷiʔəxʷ čəd ləšuuc

xʷiʔ=əxʷ čəd lə=šuɬ–c

NEG=now 1SG.SUB NEGP=see–ALTV

‘I am not to look at it’

(Hess 1998: 87, line 250)

b. xʷiʔ č$xʷ b$l$čaʔkʷ

xʷiʔ č$xʷ b$=l$=čaʔkʷ

NEG 2SG.SUB ADD=NEGP=seaward

‘don’t you come down to the shore again!’

[JS Basket Ogress, line 37]

c. ƛ’ubəxʷ čələp xʷiʔ ləbəgʷəlald

ƛ’ub=əxʷ čələp xʷiʔ lə=bə=gʷəlal–d

well=now 2PL.SUB NEG NEGP=ADD=hurt–ICS

‘you folks had better not punish him any more’

(Hess 2006: 73, line 748)

Note the different relative ordering of the additive and negative proclitics in (643b) in

comparison with the example in (643c). As with other cases of variable ordering of clitics

(Section 2.8), the conditions on the order of the proclitics are as yet undetermined.

In (643c), the negative adverb is preceded by another adverb, ƛ’ub ‘well’. In such cases, pre-

predicate particles and clitics follow the sentence-initial adverb and precede xʷiʔ. Although

lexical adverbs are usually ordered freely with respect to one another, xʷiʔ seems to consistently

follow other predicate-modifying words when these appear in the same clause. This is likely due

to the fact that xʷiʔ can take scope over another adverb that immediately follows it; ensuring that

xʷiʔ follows any other adverb it does not take scope over thereby avoids a potential structural

ambiguity between clauses like that in (643c), where the negative takes scope only over the

verbal predicate, and clauses like those in (644), where the negative actually does takes scope

over the other adverb:

(644) a. x%ɬ ti xʷiʔ ləhaʔaʔɬ šuɬ sɬadəyʔ tsiʔiɬ həbuʔ

xə%ɬ ti xʷiʔ lə=haʔ–aʔɬ šuɬ sɬadəyʔ tsiʔiɬ həbuʔ

seemingly NEG NEGP=DSTR see woman PROX:FEM pigeon

‘it seems that Pigeon was a not-good looking woman’

(Hess 2006: 22, line 9)

468

b. huy ʔusaʔil əw’ə sixʷ tiʔəʔ xʷiʔ ləhaʔɬ qaw’qs, ʔəxʷk’ʷəlx%qs

huy ʔu–saʔ–il əw’ə sixʷ tiʔəʔ xʷiʔ lə=haʔɬ qaw’qs ʔəxʷk’ʷəlx%qs

SCONJ PFV–bad–INCH PTCL PTCL PROX NEG NEGP=good raven flared.nostrils

‘then this no-good, flared-nostril Raven got into trouble’

(Hess 1998: 63, line 82)

In (644a), the negative adverbial has scope over the adverb haʔɬ ‘good’, and the resulting phrase

xʷiʔ ləhaʔɬ takes scope over √šuɬ ‘appear, be visible’. The expression in (644b), a modifying

phrase contained inside an NP (see Section 7.4.1 above), xʷiʔ ləhaʔɬ ‘no-good’ is a common

idiomatic expression found throughout the texts and is used to refer to a base, mean, or

disreputable person.

Another indicator of the scope of xʷiʔ in adverbial negatives is the negative proclitic lə=,

which attaches itself directly to the negated element, as illustrated in (645):

162

(645) gʷəl xʷiʔ ləlaʔb ʔaciɬtalbixʷ

gʷəl xʷiʔ lə=laʔb ʔaciɬtalbixʷ

SCONJ NEG NEGP=really person

‘so he was not really a person’

[DS Star Child, line 220]

Here, the negative adverb takes scope over the adverbial laʔb ‘really’ rather than over the main

predicate of the clause ʔaciɬtalbixʷ ‘person’ (that is, the sentence means ‘he was not-really a

person’ rather than ‘he was really a non-person’), and so the negative proclitic appears on laʔb

and not on ʔaciɬtalbixʷ. This is in direct contrast with the sentence in (646), where the scope of

negation includes the predicate nominal, and the negative proclitic appears on the noun itself:

(646) a. xʷiʔ čəd ləʔaciɬtalbixʷ

xʷiʔ čəd lə=ʔaciɬtalbixʷ

NEG 1SG.SUB NEGP=person

‘I am not an Indian’

(Kroeber 1999: 158)

162

This element has occasionally been glossed as the progressive prefix in earlier literature; however, it is shown to

be a clitic by its position preceding the additive clitic bə= in (643c) above, and its semantics clearly separate it from

the homophonous aspectual prefix.

469

As indicated by the full gloss of this sentence, the word ʔaciɬtalbixʷ ‘person’ — like its

equivalents in many North American languages — can be used to distinguish aboriginals or

people of in-group origin from outsiders (in this case, people of European descent).

The example in (646) also illustrates another property of the negative adverb, its ability to

negate non-verbal predicates. This applies not only to nouns, as in (646) and (647a), but also to

s=nominals (Section 7.4.2.1), as in (647b):

(647) a. xʷiʔ ləpišpiš tiʔiɬ

xʷiʔ lə=pišpiš tiʔiɬ

NEG NEGP=cat DIST

‘that is not a cat’

(Hess 1995: 94, ex. 5)

b. xʷiʔ lədsgʷaʔ

xʷiʔ lə=d=s=gʷaʔ

NEG NEGP=1SG.PO=NM=one’s.own

‘(they) are not mine’

(Hess 1998: 85, line 200)

The effect of applying the negative adverb to a nominal predicate is to create an expression

negating the inclusion of the referent of the subject in the class denoted by the predicate nominal,

or to negate its equivalence to the referent of the predicate nominal, as in (648b)

(648) a. diɬ sdaʔs, gʷəl ʔaciɬtalbixʷ, həlaʔb sdaʔs tsiʔiɬ k’aʔk’aʔ

diɬ sdaʔ–s gʷəl ʔaciɬtalbixʷ həlaʔb sdaʔ–s

FOC name–3PO belong.to person truly name–3PO

tsiʔiɬ k’aʔk’aʔ

DIST:FEM crow

‘that is her name, belonging to the Native people, the true name of Crow’

b. xʷiʔ lək’aʔk’aʔ

xʷiʔ lə=k’aʔk’aʔ

NEG NEGP=crow

‘it (i.e., her true name) is not “Crow”’

(Hess 1998: 57, lines 23–24)

As is often the case with nominal predicates, the exact relationship between the subject and the

predicate being asserted, or negated, is often dependent on context.

470

The negative adverb can also be used to negate other types of non-verbal predicates such as

adverbs, as shown in (649):

(649) a. xʷiʔ ləhaʔkʷ tiʔəʔ səsɬaɬlils

xʷiʔ lə=haʔkʷ tiʔəʔ s=ʔəs–ɬaɬlil=s

NEG NEGP=long.time PROX NM=STAT–live.together=3PO

‘they had not been living together for long’

(Hess 2006: 3, line 6)

b. gʷəɬčis čəxʷ ʔal tiʔəʔ xʷiʔ ləlil čxʷa cuuc

gʷə=ɬčil–s čəxʷ ʔal tiʔəʔ xʷiʔ lə=lil čxʷa cut–c

SBJ=arrive–ALTV 2SG.SUB at PROX NEG NEGP=far 2SG.COORD say–ALTV

‘you can reach him at (a place that is) not far from here and speak to him’

(Hess 2006: 28, line 49)

Adverbial negation of non-verbal predicates is not restricted to matrix clauses, but is also found

in various types of subordinate clause, as in the adverbial prepositional phrase headed by ʔal ‘at’

in (649b).

Not unexpectedly, xʷiʔ acting as an adverbial negator is also found in subordinate

subjunctive clauses, such as that in (650):

(650) tucutəb čəd ɬuxʷiʔəd lədᶻalqʷus

tu=cut–t–b čəd ɬu=xʷiʔ=əd lə=dᶻalqʷ•us

PAST=speak–ICS–PASS 1SG.SUB IRR=NEG=1SG.SBJ NEGP=turn•head

‘I have been told not to look over my shoulder’

(Hess 1998: 82, 116)

Note that in this example the negative adverb attracts both the irrealis proclitic and the enclitical

subjunctive subject-marker. The negative proclitic, however, remains associated with the verbal

predicate of the embedded clause, indicating the scope of negation.

8.6.2 Existential negatives

Existential negatives are propositions that negate the existence of something. Although the

syntactic structure of these expressions in Lushootseed is somewhat obscure, it seems like the

best way to analyze them is as impersonal subjectless constructions whose predicate is the

471

negative adverbial which in turn takes a nominal predicate complement bearing the subjunctive

clitic:

(651) a. xʷiʔ gʷəstutubš

xʷiʔ gʷə=stu–tubš

NEG SBJ=ATTN–man

‘there were no boys’

[LA Basket Ogress, line 119]

b. xʷiʔ gʷəstabəxʷ

xʷiʔ gʷə=stab=əxʷ

NEG SBJ=what=now

‘there is nothing (left)’

[ML Mink and Tutyika II, line 101]

Quite frequently, the complement in these constructions is introduced by the remote/hypothetical

determiner, kʷi:

(652) a. xʷiʔ kʷi gʷəpišpiš

xʷiʔ kʷi gʷə=pišpiš

NEG REM SBJ=cat

‘there are no cats’

(Hess 1995: 95, ex. 3)

b. xʷiʔ kʷi gʷəbiac kʷi gʷəstab

xʷiʔ kʷi gʷə=biac kʷi gʷə=stab

NEG REM SBJ=meat REM SBJ=what

‘there is no meat or anything’

(Hess 1998: 88, line 274)

c. xʷiʔ uʔxʷ kʷi gʷəstab ʔə tiʔiɬ dəxʷuxʷiʔxʷiʔs

xʷiʔ uʔxʷ kʷi gʷə=stab ʔə tiʔiɬ dəxʷ=ʔu–xʷiʔxʷiʔ=s

NEG PTCL REM SBJ=what PR DIST ADNM=PFV–forage=3PO

‘he still didn’t have anything from his fishing’

(Hess 2006: 14, line 85)

Other than negating the existence of something, one of the principal uses of the construction with

kʷi is the negation of possession:

(653) xʷiʔ kʷi gʷədsqʷəbayʔ

xʷiʔ kʷi gʷə=d–sqʷəbayʔ

NEG REM SBJ=1SG.PO–dog

‘I don’t have a dog’

(Hess 1998: 79, line 35)

472

In these contexts, the subjunctive complement of the negative predicate is a possessed noun, the

utterance having a literal gloss along the lines of ‘there is nothing that would be my dog’. The

exact semantic contribution of the determiner in the constructions in (652) and (653) is difficult

to ascertain; however, in texts the construction with kʷi tends to appear in contexts where the

non-existent object is something that was expected to exist or that was being looked for, whereas

the constructions without the determiner in (651) are merely observations of something’s

absence. This would be consistent with the use of kʷi in expressions denying possession of

something, as such utterances would normally be restricted to contexts where negated item is

something the speaker is expected to possess.

As noted above, the syntactic structure of the existential negative is rather obscure,

particularly in constructions such as those in (652) and (653) where the NP following the

negative predicate is introduced by a determiner and so potentially could be analyzed a syntactic

predicate. There are, however, a number of advantages to analyzing the following NP as a

predicate complement. The first of these is that it allows us to treat both types of existential

negatives as the same type of syntactic construction, given that the absence of a determiner from

expressions such as those in (651) militates against its interpretation as a syntactic subject. This

analysis also allows for a clear distinction between the negative adverb used as a syntactic

predicate with the meaning ‘not be there’ in (638) above, and used as the impersonal existential

negator with the meaning ‘there is no’ shown in (651). Furthermore, treating the subjunctive

element as a complement — and, hence, as part of the predicate — also avoids a violation of the

Lushootseed constraint that subject be Thematic (Section 11.1), given that utterances such as

those in (651) are, like their English counterparts, all-Rhematic. Evidence from the

nominalization of existential negative clauses also seems to back up this analysis:

David Beck 10-2-7 2:19 PM

Comment: note that there is no subjunctive

clitic here. see also line 3 of MW Star Child

473

(654) ʔux%ʷcəbəxʷ ʔə sp’ic’ikʷ tiʔiɬ sqas ʔə tiʔiɬ sxʷiʔs kʷi gʷat ləšəqəd

ʔux%ʷ–c–b=əxʷ ʔə sp’ic’ikʷ tiʔiɬ sqa–s

go–ALTV–PASS=now PR Diaper.Child DIST older.brother–3PO

ʔə tiʔiɬ s=xʷiʔ=s kʷi gʷat lə=šəq–d

PR DIST NM=NEG=3PO REM who NEGP=be.high–ICS

‘Diaper Child went to his older brother when there was no one to put him up high’

[DS Star Child, line 268]

In this example, the negative clause xʷiʔ kʷi gʷat ləšəqəd ‘there is no one to put him up high’ is

nominalized and serves as the complement of the preposition ʔə in an adverbial prepositional

phrase (Section 9.5); the resulting s=nominal (Section 7.4.2.1) takes the third-person possessive

subject clitic =s. If kʷi gʷat ləšəqəd were the subject of the nominalized clause, we would have

expected the third-person clitic to be absent and for kʷi gʷat ləšəqəd to be introduced by the

preposition ʔə. As this is not the case, its treatment as a complement seems to be the most

straightforward option.

163

8.6.3 Non-future negatives

When negating events in the past or the present, Lushootseed uses a construction very like

the existential negative construction shown in (651) above, in which the negative adverbial acts

as an impersonal negative predicate, taking a bare complement marked with the subjunctive

clitic, only in these cases the complement is an s=nominal rather than a lexical noun

(655) a. xʷiʔ uʔxʷ gʷəsɬaʔ ʔə tiʔəʔ čaləs

xʷiʔ uʔxʷ gʷə=s=ɬaʔ ʔə tiʔəʔ čaləs–s

NEG PTCL SBJ=NM=arrive PR PROX hand–3PO

‘his hand still can not reach it’

(Hilbert & Hess 1977: 23)

b. xʷiʔəxʷ gʷəsx&aabs dxʷʔal sɬčil ʔə tsiʔəʔ bədaʔs

xʷiʔ=əxʷ gʷə=s=x%aab=s dxʷ–ʔal s=ɬčil ʔə tsiʔəʔ bədaʔ–s

NEG=now SBJ=NM=cry=3PO CNTRPT–at NM=arrive PR PROX:FEM child–3PO

‘he isn’t crying (even) when her daughter arrives’

[HM Star Child, line 48]

163

It should be noted, however, that this is far from definitive evidence, as the =s plus non-oblique NP realization of

the subject of nominalized clauses is attested for some clause types (see Section 7.4.2.1).

474

c. xʷiʔ gʷəƛ’usuʔidigʷat ʔə tsiʔəʔ waq’waq’

xʷiʔ gʷə=ƛ’u=s=ʔu–ʔidigʷat ʔə tsiʔəʔ waq’waq’

NEG SBJ=HAB=NM=PFV–say.what PR PROX:FEM frog

‘Frog would not say anything intelligible’

(Hess 2006: 3, line 17)

The effect of these constructions is to negate the realization of the event expressed by the

s=nominal. On the whole, this is the most common negative expression found in the textual

corpus, although it has been neglected in the literature on negation in Lushootseed (e.g., Hess

1995, Davis 2005).

164

This negative construction is not restricted to matrix clauses, but is also found in subordinate

subjunctive clauses such as those in (656):

(656) a. qəldub gʷəxʷiʔəs kʷi suʔukʷukʷs ɬuɬax%iləs

qəl–dxʷ–b gʷə=xʷiʔ=əs kʷi s=ʔu–ʔukʷukʷ=s ɬu=ɬax%–il=əs

stopped–DC–PASS SBJ=NEG=3SBJ REM NM=PFV–play=3PO IRR=dark–INCH=3SBJ

‘he was told not to play when it was dark’

[DM Basket Ogress, line 3]

b. gʷətulək’ʷəd bək’ʷ tiʔəʔ wiw’su gʷəxʷiʔəs tiʔəʔ sgʷəlaltəbs

gʷ$=tu=l$k’ʷ–d b$k’ʷ tiʔ$ʔ wiw’su gʷ$=xʷiʔ=$s tiʔ$ʔ

SBJ=PAST=eaten–ICS all PROX children SBJ=NEG=3SBJ PROX

s=gʷəlal–t–b=s

NM=harmed–ICS–PASS=3PO

‘she would have eaten all the children if she hadn’t been killed’

[ML Basket Ogress, line 111]

164

Perhaps the primary reason for this arises from a transcription practice followed in the Lushootseed Readers

(Hess 1995, 1998, 2006), whereby the remote determiner kʷi was consistently added to nearly all of the instances of

negated s=nominals, although it is not heard on the tapes. The bulk of these transcriptions were made by Thom Hess

with the help of Levi Lamont and may represent a form of hyper-correction; alternatively, it may have been done for

pedagogical reasons. The pattern with kʷi does exist but, based on the patterns found in other texts (including texts

from the principal speaker in the Lushootseed Readers, Martha Lamont, transcribed at a later date), it appears to be a

separate construction (Section 8.6.4) and many of the amended forms in the Readers appear to have been correct as

spoken in the recordings, without the determiner.

475

c. gʷətuʔaʔa uʔxʷ gʷəxʷiʔəs tiʔəʔ tusgʷəlaltəbs tsiʔiɬ sxʷəyuq’ʷ ʔi tsiʔiɬ ʔay’əds

gʷ$=tu=ʔa–ʔa uʔxʷ gʷ$=xʷiʔ=$s tiʔ$ʔ

SBJ=PAST=INTNS–be.there PTCL SBJ=NEG=3SBJ PROX

tu=s=gʷəlal–t–b=s tsiʔiɬ sxʷəyuq’ʷ ʔi tsiʔiɬ

PAST=NM=harmed–ICS–PASS=3PO DIST:FEM Basket.Ogress and DIST:FEM

ʔay’əd–s

companion–3PO

‘they would be right there if they hadn’t killed the Basket Ogress and her friend’

[ML Basket Ogress, line 186]

The first example in (656a) is a negative jussive clause in which the negation is of a future

hypothetical event. In the next two clauses, the negative construction is used to express

counterfactual condition.

This negative construction can also be nominalized, either by s=, as in (657),

(657) a. ɬuhuyil sq’axʷ dxʷʔal gʷəsxʷiʔs gʷəšədᶻaldubuts ʔə kʷi haʔkʷ

ɬu=huy–il s–q’axʷ dxʷʔal gʷə=s=xʷiʔ=s

IRR=be.done–INCH NP–frozen CNTRPT–at SBJ=NM=NEG=3PO

gʷə=s=šədᶻal–dxʷ–but=s ʔə kʷi haʔkʷ

SBJ=NM=go.outdoors–DC–REFL=3PO PR REM long.time

‘it will become ice so that he can not get himself out of the house for a long time’

(Hess 1998: 101, line 269)

b. ləkiis dxʷʔal tiʔiɬ sxʷiʔs gʷəgʷədils

lə=kiis dxʷ–ʔal tiʔiɬ s=xʷiʔ=s gʷə=gʷəd–il=s

PROG=stand CNTRPT–at DIST NM=NEG=3PO SBJ=down–INCH=3PO

‘he stood so as not to be underneath’

[MS Basket Ogress, line 19]

or by dəxʷ= as in (658),

(658) yəx%i čəd ʔəsʔi tə dəxʷiʔs gʷədsʔilid kʷi ʔəsʔuʔux%ʷ syəyəhub

yəx%i čəd ʔəsʔi tə dəxʷ=xʷiʔ=s gʷəd=s=ʔili–d

because 1SG.SUB slightly.ill NSPEC ADNM=NEG=3PO SBJ=1SG.PO=NM=recite–ICS

kʷi ʔəs–ʔu–ʔux%ʷ syəyəhub

REM STAT–ATTN–go legend

‘because I am indisposed is why I did not recite a traditional story which goes on and on’

[HM Star Child, line 195]

Note that in both (657) and (658) the embedded negative is marked with the possessive subject

clitic =s and the following s= or dəxʷ=nominal has no determiner, nor is it introduced by the

476

preposition ʔə, as would be expected if these elements were subjects rather than complements of

the negative adverbial predicate.

8.6.4 Future negatives

While bare s=nominal complements are used in the general negation of events in the past

and the present, negation of future events — that is, the assertion that some particular event or

type of event will not occur — requires the introduction of the remote determiner kʷi, with or

without the subjunctive clitic. When used without the subjunctive clitic, the negation is of a

future event that is considered more or less definite or specific (that is, as a plausible or

foreseeable potentiality):

(659) a. xʷiʔ kʷi suʔux%ʷ čəɬ dxʷʔal tiʔəʔ ʔiq’ʷ šəgʷɬ

xʷiʔ kʷi s=ʔu–ʔux%ʷ čəɬ dxʷ–ʔal tiʔəʔ ʔiq’ʷ šəgʷɬ

NEG REM NM=PFV–go 1PL.SUB CNTRPT–at PROX swept path

‘we will not go on the cleared path’

[AW Basket Ogress, line 38]

b. gʷəl tudxʷcutəb həlgʷəʔ, ʔuˑ xʷiʔəxʷ kʷi ʔahəxʷ gʷəčaldubuɬ

gʷəl tu=dxʷcutəb əlgʷəʔ ʔu xʷiʔ=əxʷ kʷi ʔa=əxʷ

SCONJ PAST=think PL INTJ NEG=now REM be.there=now

gʷə=čal–dxʷ–buɬ

SBJ=chased–DC–1PL.OBJ

‘and then they thought, oh, there will be nothing that can catch us’

(Hilbert and Hess 1977: 16)

c. xʷiʔ kʷi stiləb čəɬ ɬut’uk’ʷ

xʷiʔ kʷi s=tiləb čəɬ ɬu=t’uk’ʷ

NEG REM NM=suddenly 1PL.PO IRR=go.home

‘we will not go home right away’

(Hess 2006: 72, line 710)

When combined with the irrealis clitic ɬu=, the complement phrase negates a future state (660a)

or a future activity (660b and c):

477

(660) a. xʷiʔəxʷ kʷi ɬubəsəshuys ʔaciɬtalbixʷ

xʷiʔ=əxʷ kʷi ɬu=bə=s=ʔəs–huy=s ʔaciɬtalbixʷ

NEG=now REM IRR=ADD=NM=STAT–be.made=3PO person

‘she (Owl) will not be a person anymore’

(Hess 2006: 9, line 155)

b. xʷiʔəxʷ kʷi ɬudsuxʷiʔxʷiʔ ʔə kʷi skʷiʔxʷ

xʷiʔ=əxʷ kʷi ɬu=d=s=ʔu–xʷiʔxʷiʔ ʔə kʷi skʷiʔxʷ

NEG=now REM IRR=1SG.PO=NM=PFV–forage PR REM bracken.fern.root

‘now I will not (have to) gather bracken fern roots’

[DS Star Child, line 164]

c. xʷiʔəxʷ kʷi ɬadsux%ʷubil ʔi tiʔəʔ biʔbədaʔ

xʷiʔ=əxʷ kʷi ɬu=ad=s=ʔu–x%ʷubil ʔi tiʔəʔ biʔ–bədaʔ

NEG=now REM IRR=2SG.PO=PFV–be.quiet CONJ PROX ATTN–offspring

‘you and your little child will never be quiet’

(Hess 2006: 8, line 140)

In each of these datasets, the negation is of an event or state in the future relative to the speech

act or the narrative present time of the utterance. The only exceptions to this generalization are

constructions such as those in (661) where the complement of xʷiʔ is the verb haydxʷ ‘know

something’:

(661) a. xʷiʔ kʷi səsaydxʷs gʷəsəsk’ʷəčcuts

xʷiʔ kʷi s=ʔəs–hay–dxʷ=s gʷə=s=ʔəs–k’ʷəč–t–sut=s

NEG REM NM=STAT–known–DC=3PO SBJ=NM=STAT–wild–DC–ICS–REFL=3PO

‘she did not know that she should be leery’

(Hess 2006: 40, line 444)

b. xʷiʔ kʷ səsaydxʷs əlgʷəʔ tiʔəʔ dəxʷɬaliltubs əlgʷəʔ ʔə tiʔiɬ x%%pay’ac sčaɬaʔs əlgʷəʔ

swatixʷtəd

xʷiʔ kʷi s=ʔəs–hay–dxʷ=s əlgʷəʔ tiʔəʔ dəxʷ=ɬalil–txʷ–b=s

NEG REM NM=STAT–known–DC=3PO PL PROX ADNM=go.ashore–ECS–PASS=3PO

əlgʷəʔ ʔə tiʔiɬ x%%pay’•ac s=čaɬaʔ=s əlgʷəʔ swatixʷtəd

PL PR DIST red.cedar NM=not.recognize=3PO PL land

‘they didn’t know where the red cedar beached them so they didn’t recognize the land’

(Hess 2006: 59, line 413)

In these cases, what is negated is whether or not the protagonist knows a particular fact, the

knowing taking place in the narrative present. Without further attestations of the non-future use

478

of such negatives with the remote determiner, this will have to be treated as an idiomatic

expression, particular to this type of complement.

The future negative with the remote determiner is also attested in a few cases with the

subjunctive clitic gʷə=. In all, there are only four of these in the corpus over and above cases

where gʷə= (or, in a few cases, kʷi) were added by editorial amendment (see footnote 164

above). Three of the four occur in contexts where a request or a refusal to comply with a

command is being softened by the use of the subjunctive clitic:

(662) a. xʷiʔ ʔu kʷi gʷadsxʷit’ilc dxʷgʷəd

xʷiʔ ʔu kʷi gʷə=ad=s=xʷit’–il–t–s dxʷ–gʷəd

NEG INT REM SBJ=2SG.PO=NM=go.down–INCH–ICS–2SG.OBJ CNTRPT–down

‘will you not lower me down?’

(Hess 2006: 29, line 162)

b. xʷiʔ xʷiʔ siʔiʔab kʷi gʷədsgʷiid

xʷiʔ xʷiʔ siʔ–iʔab kʷi gʷə=d=s=gʷi–d

NEG NEG PL–noble REM SBJ=1SG.PO=NM=call–ICS

‘no, no, noble sirs, I won’t call them’

(Hess 1998: 79, line 32)

The second sentence here, spoken by Pheasant, a humble and respectful soul addressing some

supernatural hunters, contrasts with the sentence in (663), spoken by the rude and deceitful

Raven to his sister, Crow:

(663) xʷiʔ gʷədsgʷiid əlgʷəʔ

xʷiʔ gʷə=d=s=gʷi–d əlgʷəʔ

NEG SBJ=1SG.PO=NM–call–ICS PL

‘I wouldn’t invite them’ (lit. ‘I’m not inviting them’)

(Hess 1998: 58: 55)

Although the sentence in (663) is framed as a negation of a past or present event (Section 8.6.3),

it is effectively a refusal by Raven to comply with Crow’s instructions (which are not in

accordance with his scheme to steal her food).

The fourth attestation of this construction occurs in a context where the negated event is not a

future event, but is rather the achievement of something that had been promised:

479

(664) xʷiʔəxʷ kʷi gʷəshuydxʷs kʷi gʷəsaxʷəbs x%ʷul’ab ʔə tiʔiɬ ƛ’uscuts

xʷiʔ=əxʷ kʷi gʷə=s=huy–dxʷ=s kʷi gʷə=s=saxʷəb=s x%ʷul’ab

NEG=now REM SBJ=NM=be.done–DC=NM REM SBJ=NM=jump=3PO thus

ʔə tiʔiɬ ƛ’u=s=cut=s

PR DIST HAB=NM=say=3PO

‘he has not been able to jump the way he was saying (he would)’

[ML Mink and Tutyika, line 69]

Unfortunately, the context of the sentence is not clear enough to give us a reliable sense of the

exact temporal and modal interpretation of the utterance (specifically, the context would also

allow an interpretation of this sentence as a future conditional — i.e., ‘he will not be able to jump

the way he says he will’), leaving us with only the tense and mood of the English translation to

go on. This is not a reliable basis on which to draw inferences about the tense and mood of the

Lushootseed sentence. Given the paucity of unequivocal examples of such sentences in the

corpus to date, a fuller understanding of future negative constructions with subjunctive

complements will have to await the uncovering of further textual attestations.

8.6.5 Negative imperatives

Negative imperatives can take a fairly wide range of forms, one of the most textually

frequent of these being that of the future negative construction described in the preceding

section:

(665) a. xʷiʔ kʷi adsdᶻalqʷusbid tiʔiɬ adsəsčəbaʔ

xʷiʔ kʷi ad=s=dᶻalqʷ•us–bi–d tiʔiɬ ad=s=ʔəs–čəbaʔ

NEG REM 2SG.PO=NM=turn–MAP–ICS DIST 2SG.PO=STAT–laden

‘do not look over your shoulder at what you have on your back!’

(Hess 1998: 81, line 103)

b. xʷiʔ kʷi sxʷaacləp tsi staləɬləp

xʷiʔ kʷi s=xʷaac=ləp tsi staləɬ=ləp

NEG REM NM=deny=2PL.PO SPEC:FEM niece=2PL.PO

‘don’t you folks deny your niece!’

(Hess 1998: 91, line 9)

480

c. xʷiʔəxʷ kʷi bəsʔux%ʷləp

xʷiʔ=əxʷ kʷi bə=s=ʔux%ʷ=ləp

NEG=now REM ADD=NM=go=2PL.PO

‘don’t you guys go again!’

(Hess 1998: 94, line 106)

The sentences in (665) are predicated on the negative adverbial xʷiʔ and take as complements

s=nominals with second-person possessive subjects introduced by the remote determiner kʷi,

making them effectively negations of a future event with a second-person protagonist

Negative imperatives are also found in the form of non-future negatives (Section 8.6.3):

(666) a. xʷiʔ gʷəsʔux%ʷləp

xʷiʔ gʷə=s=ʔux%ʷ=ləp

NEG SBJ=NM=go=2PL.PO

‘don’t you guys go!’

[AJ Basket Ogress, line 5]

b. xʷiʔ gʷadsq’puc ʔə kʷi gʷəstab

xʷiʔ gʷə=ad=s=q’pu–t–s ʔə kʷi gʷə=stab

NEG SBJ=2SG.PO=NM–pay–ICS–1SG.OBJ PR REM SBJ=what

‘don’t pay me anything!’

(Hess 2006: 30, line 191)

These expressions have the form of the negation of a non-future event expressed as an

s=nominal in the subjunctive mood with a second-person subject. What exactly the semantic or

pragmatic difference between the forms in (665) and those in (666) are is not apparent from the

contexts in which the attestations are found and so will have to remain an open question until

more informative examples are uncovered in texts.

There are also one or two examples in the corpus of negative imperatives that take the form

of an ordinary adverbial negative in the indicative mood with a second-person subject:

(667) xʷiʔ čəxʷ bələčaʔkʷ kʷsi səsƛ’ayayəlqəb

xʷiʔ čəxʷ bə=lə=čaʔkʷ kʷsi səsƛ’ayayəlqəb

NEG 2SG.SUB ADD=NEGP=seaward REM:FEM monster

‘don’t you come down to the shore again, monstress!’

(lit. ‘you won’t come down to the shore again, monstress’)

[JS Basket Ogress, line 38]

481

This construction, like those shown in (665)and (666), is identical to the ordinary negative

expression of its corresponding type, and any imperative force it has is apparently drawn only

from context. This is in keeping with the general behaviour of imperatives in Lushootseed, which

are frequently expressed as simple declarative sentences with second-person subjects (see

Section 8.5 above).

In addition to the negative imperative types shown above, Lushootseed does have a special

construction that might be characterized as a negative causative imperative, used to express

negative commands or admonishments. These are based on the verb xʷiʔtxʷ, which normally

means ‘cause not to be there’, but in this case forms a complex predicate with another verb

meaning ‘V’ to form an expression meaning ‘not let V happen’, as in the following examples:

(668) a. xʷiʔtxʷ ləbakʷɬ

xʷiʔ–txʷ–Ø Ø lə=bakʷɬ

NEG–ECS–3OBJ 2SG.IMP NEGP=be.hurt

‘don’t let him get hurt!’

b. ie xʷiʔtubš ɬi ləxʷit’il

ie xʷiʔ–txʷ–bš ɬi lə=xʷit’il

INTJ NEG–ECS–1SG.OBJ 2PL.IMP NEGP=descend

‘don’t you folks let me fall!’

(Bates, Hess & Hilbert 1994: 252)

In these sentences, the verb xʷiʔtxʷ takes as a complement a verb bearing the negative proclitic

lə=. As seen in (668b), the subject of the verbal complement becomes the direct object of

xʷiʔtxʷ, whose subject is a second-person imperative subject-marker (Section 8.5).

Unfortunately, this construction is not found in the present corpus, these de-contextualized

examples from the Lushootseed Dictionary and a handful of sentences in Hess & Hilbert (1976),

a pedagogical grammar, being the only attestations available to date.

165

The Lushootseed Dictionary also gives the following sentence:

482

8.7 Desideratives

Periphrastic desiderative constructions are based on the verb x&aƛ’txʷ ‘want something, desire

something’ (morphological desideratives are discussed in Section 2.1.1.7). Like the English verb

want, x&aƛ’txʷ can be used as an ordinary transitive verb, expressing a desire for a particular

object or thing:

(669) a. x%aƛ’txʷ kʷi sčəlus

x%aƛ’–txʷ kʷi sčəlus

desire–ECS REM fish.tips

‘he wanted fish tips’

[AJ Basket Ogress, line 26]

b. x%ʷul’əxʷ čəxʷ ʔukʷədad, gət, taʔa gʷəx%aƛ’txʷəxʷ, kʷi gʷəɬadst’uk’ʷtx

x%ʷul’=əxʷ čəxʷ ʔu–kʷəda–d gət taʔa gʷə=x%aƛ’–txʷ=əxʷ

only=now 2SG.SUB PFV–taken–ICS guy UNQ:DMA SBJ=desire–ECS=now

kʷi gʷə=ɬu=ad=s=t’uk’ʷ–tx

REM SBJ=IRR=2SG.PO=NM=go.home–ECS

‘you just take it, guy, if you want it, you should take it home’

(Hess 1998: 73, 191)

In these cases, the verb is an ordinary transitive stem and takes an NP object. In a desiderative

construction, the same verb is used, but it takes an s-nominal complement expressing the desired

event. The complement, generally in the subjunctive mood, is optionally introduced by the

remote determiner kʷi:

(i) xʷiʔtxʷ gʷəl ʔəstagʷəxʷ tadbədaʔ

xʷiʔ –txʷ gʷəl ʔəs–tagʷəxʷ ti ad=bədaʔ

NEG–ECS SCONJ STAT–hunger SPEC 2SG.PO=offspring

‘do not let your son go hungry!’

(Bates, Hess & Hilbert 1994: 252)

This sentence apparently lacks the negative proclitic; however, the presence of the conjunction gʷəl at this place in

the sentence seems unusual, and it is possible that what was actually said was gʷəl(ə)ʔəstagʷəxʷ, the verb bearing

the subjunctive and the negative proclitics. This will have to remain an open question pending a closer phonetic

examination of the audio recordings when these become available in digitized form.

483

(670) a. x%aƛ’tub ʔə tiʔəʔ sqʷəbqʷəbayʔ (kʷi) gʷəsx%ʷuƛ’utəbs

x%aƛ’–txʷ–b ʔə tiʔəʔ sqʷəb–qʷəbayʔ kʷi gʷə=s=x%ʷuƛ’u–t–b=s

desire–ECS–PASS PR PROX DSTR–dog REM SBJ=NM=desire–ICS–PASS=3PO

‘the dogs want that he get chewed up’

(lit. ‘it is desired by the dogs that he be chewed up’)

(Hess 1998: 46)

b. x%aƛ’txʷ čəɬ kʷi gʷadst’ilib, gʷadsdᶻubalikʷ dᶻixʷbid ʔə kʷi ɬadsq’ʷəltubuɬ

x%aƛ’–txʷ čəɬ kʷi gʷə=ad=s=t’ilib gʷə=ad=s=dᶻubalikʷ

desire–ECS 1PL.SUB REM SBJ=2SG.PO=NM=sing SBJ=2SG.PO=NM=dance

dᶻix–bid ʔə kʷi ɬu=ad=s=q’ʷəl–t–ubuɬ

first–RNLN PR REM IRR=2SG.PO=NM=cooked–ICS–1PL.OBJ

‘we want you to sing and dance before you cook us’

[AJ Basket Ogress, line 84]

c. huy gʷəl x%aƛ’tubəxʷ ʔə tsiʔəʔ wiw’su kʷi gʷəsuʔx%ʷs, gʷəsɬaʔs

huy gʷəl x%aƛ’–txʷ–b=əxʷ ʔə tsiʔəʔ wiw’su kʷi

SCONJ SCONJ desire–ECS–PASS=now PR PROX:FEM children REM

gʷə=s=ʔu–ʔux%ʷ=s gʷə=s=ɬaʔ=s

SBJ=NM=PFV–go=3PO SBJ=NM=arrive=3PO

‘the children wanted to go there, to arrive’

(lit. ‘going there, arriving, was desired by the children’)

[AJ Basket Ogress, line 2]

The optionality of the determiner — indicated explicitly in (670a) and its absence on second of

two sequential s-nominals in (670b) and (c) — indicates that the s-nominal is functioning as a

complement of the verb rather than as its direct object. As shown by (670b), the protagonist of

the desired action need not be the one who desires its realization; note also that in (670a),

although the AGENTS of the desired action are the desirers, the s-nominal complement is in the

passive voice and the expression of the AGENT is suppressed.

Although the nominalized complement of x&aƛ’txʷ appears more frequently in the subjunctive

mood, there are examples in the corpus of complements in the indicative mood introduced

simply by the remote determiner:

166

There are no examples in the current corpus of complements in the indicative mood without the remote

determiner.

484

(671) a. x%aƛ’txʷ əlgʷəʔ kʷi səsq’ʷuʔləpəxʷ

x%aƛ’–txʷ əlgʷəʔ kʷi s=ʔəs–q’ʷuʔ=ləp=əxʷ

desire–ECS PL REM NM=STAT–gather=2PL.PO=now

‘they want you guys to gather’

(Hess 2006: 73, line 737)

b. tux%ʷ čəɬ x%aƛ’txʷ kʷi sƛ’ubləp ʔəsqʷib …

tux%ʷ čəɬ x%aƛ’–txʷ kʷi s=ƛ’ub=ləp ʔəs–qʷib

only 1PL.SUB desire–ECS REM NM=well=2PL.PO STAT–prepared

‘we just want you guys to be ready’

(Hess 2006: 74, line 759)

Without further attestations or the assistance of a native speaker, it is difficult to be sure what the

difference between the two constructions is; based on the few examples available, however, the

distinction seems to be one of control over the outcome of the event, the uncertainty of the

subjunctive mood being associated with diminished control. Both the examples in (671) occur in

a context where the action (or its outcome) is wished for both by the desirers and the actors who

are to carry out the desire; thus, the desire is more likely to be realized, resulting in greater

determinacy and control over the event on the part of the desirers. In (670a), the outcome of the

desired action (being chewed on) is not wished for by the one who is to be chewed on, is likely

to be resisted, and, in fact, ends up being thwarted by the dogs’ owners, while in (670b), the

desirers are children about to be eaten by the Basket Ogress and begging of her a favour that she

may or may not consent to grant. In (670c) the desirer and the potential actor are one and the

same (the children), but the sentence occurs in a context where the children’s desired action has

been forbidden by their parents, whose control over the situation is thus diminished by their

children’s recklessness (which eventually leads to the predicament in 670b). While these are

somewhat heterogeneous situations, they seem to follow the general patterns shown by the

diminished control suffix (Section 2.1.2.3) in Lushootseed and the overall semantics of lack of or

diminished control in other Salishan languages.

The verb x&aƛ’txʷ is based, at least etymologically, on the bound radical º√x&aƛ’ ‘desire’. On

its own, this radical is attested only in nominalized form in the word sx&aƛ’ ‘desired one’, which

485

is used like any noun as either a syntactic argument (672a) or as a syntactic predicate in

expressions of desiring or liking, as in (672b) and (c):

(672) a. ʔuˑ, diɬ kʷi səsɬax%il kʷi dsx%aƛ’

ʔu diɬ kʷi s=ʔəs–ɬax%–il kʷi d=s–x%aƛ’

INTJ FOC REM NM=STAT–dark–INCH REM 1SG.PO=NP–desire

‘oh, the one that I desire is the dark one’

[DS Star Child, line 384]

b. huy sx%aƛ’s tiʔəʔ tusč’istxʷs

huy s–x%aƛ’–s tiʔəʔ tu=sč’istxʷ–s

SCONJ NP–desire–3PO PROX PAST=husband–3PO

‘well, the one she likes is her late husband’

(Hess 2006: 33, line 285)

c. sx%aƛ’s kʷi gʷəsx%ʷuƛ’utəbs

s–x%aƛ’–s kʷi gʷə=s=x%ʷuƛ’u–t–b=s

NP–desire–3PO REM SBJ=NM=chewed–ICS–PASS=3PO

‘s/he/they want that s/he/they get chewed up’

(lit. ‘it is his/her/their desire that s/he/they get chewed up’)

(Hess 1998: 46)

As in (670a), the optionality of the remote determiner in (672c) indicates that the s-nominal is

likely a complement of the nominalized radical rather than the syntactic subject of the sentence.

The conditions governing the presence or absence of the subjunctive clitic on the complement

are likely to be the same as those governing the presence of gʷə= on the complement of

x&aƛ’txʷ.

167

The negation of expressions of this type takes the form of an existential negative (Section 8.6.2) based on the

negative predicate xʷiʔ:

(i) xʷiʔəxʷ gʷədsx%aƛ’ kʷi gʷəbadsuʔəƛ’, gʷadsuʔabaqəd

xʷiʔ=əxʷ gʷə=d=s–x%aƛ’ kʷi gʷə=bə=ad=s=ʔu–ʔəƛ’

NEG=now SBJ=1SG.PO=NP–desire REM SBJ=ADD=2SG.PO=NM=PFV–come

gʷə=ad=s=ʔu–ʔabaq–d

SBJ=2SG.PO=NM=PFV–return–ICS

‘I don’t want you to come (back here), (nor for) you to return her’

(lit. ‘it is not my desire that you come back here or that you return her’)

(Hess 2006: 35, line 335)

The syntax of these expressions is potentially interesting, but there only a single attestation in the present corpus.

486

In a number of cases, sx&aƛ’, in spite of the fact that it has been nominalized, is attested with

the stative aspectual prefix:

(673) a. həlaʔb čəxʷ dəsx%aƛ’

həlaʔb čəxʷ d=s=ʔəs–x%aƛ’

really 2SG.SUB 1SG.PO=NM=STAT–desire

‘I really want you’

b. dəsx%aƛ’ kʷi dšudubicid

d=s=ʔəs–x%aƛ’ kʷi d=šuɬ–dxʷ–bicid

1SG.PO=NM=STAT–desire REM 1SG.PO=see–DC–2SG.OBJ

‘I want to see you’

(Bates, Hess & Hilbert 1994: 258)

The aspectual prefixes are otherwise limited to verbs and questions words (Section 6.1).

The radical º√x&aƛ’ is also the base for other expressions of liking or desiring. One of these,

x&aƛ’ildxʷ ‘like someone, desire someone’ is formed by an idiosyncratic combination of the

inchoative suffix -il and the diminished control suffix, -dxʷ:

(674) a. ʔux%aƛ’ildxʷ čəd

ʔu–x%aƛ’–il–dxʷ čəd

PFV–desire–INCH–DC 1SG.SUB

‘I got to liking him’

b. x%aƛ’ildxʷ tiʔiɬ ləliʔ sɬəɬadəyʔ

x%aƛ’–il–dxʷ tiʔiɬ ləliʔ sɬə–ɬadəyʔ

desire–INCH–DC DIST different PL–woman

‘he wants different women’

c. ʔacəc tiʔiɬ tux%aƛ’ildəgʷəl

ʔacəc tiʔiɬ tu=x%aƛ’–il–dxʷ–agʷəl

exist DIST PAST=desire–INCH–DC–RCP

‘there was a couple who fell in love’

(Bates, Hess & Hilbert 1994: 258)

168

The second derivation based on º√x&aƛ’ is formed with the internal causative suffix, -t:

168

This form is analyzed in the original source as x&aƛ’–il–d–agʷəl, with the internal causative transitivizing the

stem rather than the diminished control suffix; either combination of affixes would create the correct surface form,

but given that (674c) is a transparent reciprocal of (674a) and (b), it seems better not to posit a new and otherwise

unattested stem, x&aƛ’ild.

487

(675) x%i diɬ tadsx%aƛ’təb ʔə tadsx%aʔx%aʔ

yəx%i diɬ tu=ad=s=x%aƛ’–t–b ʔə ti ad–sx%aʔx%aʔ

because FOC PAST=2SG.PO=NM=desire–ICS–PASS PR SPEC 2SG.PO–in.law

‘because it was your in-law who favoured you’

(Hess 2006: 32, line 243)

There is only a single attestation of this stem, sx&aƛ’əd in the corpus to date and it is not listed in

the Lushootseed Dictionary, so little can be said about its precise meaning or in what ways it

differs from the other forms based on the same radical.

8.8 Comparatives

Comparative constructions are expressions which present the relative values of two objects

for some scalar predicate or basis of comparison. When the two objects have different values for

this predicate, the expression is a comparison of inequality, and when they have the same value,

the expression is a comparison of equivalence. In Lushootseed, a comparison of inequality

realizes the object of comparison or comparee as a syntactic subject and uses the partitive form

of the expression of the basis of comparison as the syntactic predicate:

169

(676) a. ʔiɬluƛ’ čəd dxʷʔal dəgʷi

ʔiɬ–luƛ’ čəd dxʷ–ʔal dəgʷi

PRTV–old 1SG.SUB CNTRPT–at you

‘I am older than you’

b. ʔiɬbəqʷ čəxʷ ʔu dxʷʔal ti adsuq’ʷaʔ

ʔiɬ–bəqʷ čəxʷ ʔu dxʷ–ʔal ti ad–suq’ʷaʔ

PRTV–fat 2SG.SUB INT CNTRPT–at SPEC 2SG.PO–younger.sibling

‘are you fatter than your younger sibling?’

(Hess & Hilbert 1976: II, 48)

c. lil čəxʷ ʔiɬsadᶻəb dxʷʔal ʔəca

lil čəxʷ ʔiɬ–sadᶻəb dxʷ–ʔal ʔəca

much 2SG.SUB PRTV–be.tall CNTRPT–at I

‘you are much taller than I’

(Hess & Hilbert 1976: II, 51)

The object being compared with the comparee, the standard of comparison, is realized in a

prepositional phrase introduced here by dxʷʔal ‘towards’, the marker of standard. The comparee

169

The terms “comparee,” “standard of comparison,” and “marker of standard” are drawn from Stassen (1984).

488

in these examples is expressed by a subject clitic, which (as is normally the case) obligatorily

appears in sentence-second position, following the sentence predicate or a sentence-initial

adverbial such as the predicate particle lil ‘much’ (etymologically-related to lil ‘be far’ — see

Section 2.5.3 above).

When the comparee is a full NP, it usually follows the sentence predicate as well (677a), but

in some cases may be preceded by the standard of comparison, as in (677b):

(677) a. ʔiɬhaac tiʔəʔ ƛ’əlayʔ dxʷʔal tiʔiɬ ƛ’əlayʔ ʔal tudiʔ

ʔiɬ–haac tiʔəʔ ƛ’əlayʔ dxʷ–ʔal tiʔiɬ ƛ’əlayʔ ʔal tudiʔ

PRTV–long PROX canoe CNTRPT–at DIST canoe at DIST.DMA

‘this canoe is longer than that canoe over there’

(Hess & Hilbert 1976: II, 48)

b. ʔiɬhaʔɬ dxʷʔal kʷi p’aƛ’aƛ’ ti dsgʷaʔ

ʔiɬ–haʔɬ dxʷ–ʔal kʷi p’aƛ’aƛ’ ti d–sgʷaʔ

PRTV–good CNTRPT–at REM worthless SPEC 1SG.PO–one’s

‘mine is better than nothing’

(Hess & Hilbert 1976: II, 49)

Although there is no textual data available to allow us to be certain of the conditions governing

this alternation, it seems likely that they are similar to those regulating the alternation in the

word-order in passive constructions (see Section 11.1), in which the subject NP is realized to the

right of the object agentive complement when the identity of the complement is part of the

sentence Rheme. In (677b), the expression ʔiɬhaʔɬ dxʷʔal kʷi p’aƛ’aƛ’ ‘better than nothing’

forms a conceptual unit and, as a whole, constitutes the Rheme of the sentence, whereas in

(677a) the Rheme is simply the fact that, of two Given and probably Topical Thematic elements

(the two canoes), one is longer than the other, the sentence Rheme being simply ʔiɬhaac

‘longer’, the basis of comparison.

When the standard of comparison is close by as opposed to more removed from the speaker,

tul’ʔal ‘from’ can be used as a marker of standard rather than dxʷʔal ‘towards’:

489

(678) a. ʔiɬx%əb tiʔiɬ wəq’əb ʔal tudiʔ tul’ʔal tiʔəʔ wəq’əb ʔə tiʔaʔ

ʔiɬ–x%əb tiʔiɬ wəq’əb ʔal tudiʔ tul’–ʔal tiʔəʔ wəq’əb

PRTV–heavy DIST box at DIST.DMA CNTRFG–at PROX box

ʔə tiʔaʔ

PR PROX:UNQ.DMA

‘that chest over there is heavier that this chest right here’

b. ʔiɬx%əb tiʔəʔ wəq’əb ʔal tiʔaʔ dxʷʔal tiʔiɬ wəq’əb ʔal tudiʔ

ʔiɬ–x%əb tiʔəʔ wəq’əb ʔal tiʔaʔ dxʷ–ʔal tiʔiɬ wəq’əb

PRTV–heavy PROX box at PROX:UNQ.DMA CNTRPT–at DIST box

ʔal tudiʔ

at DIST.DMA

‘this chest right here is heavier than that chest over there’

(Hess & Hilbert 1976: II, 118)

The same is true when the speaker is the standard of comparison:

(679) a. ʔiɬluƛ’ čəxʷ tul’ʔal ʔəca

ʔiɬ–luƛ’ čəxʷ tul’–ʔal dəgʷi

PRTV–old 2SG.SUB CNTRFG–at I

‘you are older than I’

b. ʔiɬluƛ’ čəd dxʷʔal dəgʷi

ʔiɬ–luƛ’ čəd dxʷ–ʔal dəgʷi

PRTV–old 1SG.SUB CNTRPT–at you

‘I am older than you’

(Hess & Hilbert 1976: II, 118)

However, Hess & Hilbert (1976) note that this deictic distinction is not consistently maintained

and dxʷʔal is frequently used simply as a default.

Hess & Hilbert (1976) also mention that there are some words which do not take the partitive

prefix in comparatives, such as the Southern Lushootseed word wələx&ʷ ‘strong’:

(680) ʔəswələx%ʷ čəxʷ tul’ ʔəca

ʔəs–wələx%ʷ čəxʷ tul’ ʔəca

STAT–strong 2SG.SUB CNTRFG I

‘you are stronger than I’

(Hess & Hilbert 1976: 49)

There is no indication of which other words follow this pattern. Note also the use of the

centrifugal directional particle tul’ (Section 2.7.4) as the marker of standard in place of

dxʷʔal/tul’ʔal in this example; this is typical of Southern Lushootseed.

490

Comparative questions soliciting the difference between a comparee and a standard take the

form of an ordinary question based on ʔəx&id ‘what happened?’ (Section 2.6.8):

(681) ʔəsʔəx%id kʷi sləliʔ ʔə tiʔiɬ haʔac dxʷʔal sx%əp’ab

ʔəs–ʔəx%id kʷi s=ləliʔ ʔə tiʔiɬ haʔac dxʷ–ʔal sx%əp’ab

STAT–what.happen REM NM=differ PR DIST horse.clam CNTRPT–at cockle

‘how does a horse clam differ from a cockle?’

(Hess & Hilbert 1976: II, 74)

In these constructions, dxʷʔal is used to introduce the standard of comparison.

In addition to the partitive construction illustrated above, there is another comparative based

on the relational suffix -bid (Section 2.2.10):

(682) a. hikʷbid ʔə tiʔəʔ sx%iʔx%aʔaʔ tiʔəʔ sʔax%ʷuʔ

hikʷ–bid ʔə tiʔəʔ sx%iʔ–x%aʔaʔ tiʔəʔ sʔax%ʷuʔ

big–RLNL PR PROX ATTN–steamer.clam PROX butter.clam

‘the butter clam is bigger than the little steamer clam’

(Hess 1998: 44, ex. 7)

b. tiʔəʔ swukʷad gʷəl hikʷbid dxʷʔal tiʔəʔ x%ʷətis

tiʔəʔ swukʷad gʷəl hikʷ–bid dxʷ–ʔal tiʔəʔ x%ʷətis

PROX loon SCONJ big–RLNL CNTRPT–at PROX silver.diver

‘as for the loon, it is bigger than the silver diver’

(Hess & Hilbert 1976: II, 73)

In these expressions, the basis of comparison is formed with the relational suffix and the standard

of comparison is contained in a prepositional phrase, the marker of comparison being either the

“empty” preposition ʔə or (as in the partitive construction) dxʷʔal. According to Hess & Hilbert

(1976), the difference between the two types of construction resides in the fact that the partitive-

based construction is focused on the difference between the comparee and the standard, whereas

the -bid construction is more a description of the comparee relative to the standard. It should be

noted, however, that it is not clear how productive the comparative construction with -bid

actually is; there are no attestations in the present corpus of -bid being use in a comparative with

any base other than hikʷ ‘big’, suggesting that this is a lexicalized idiomatic expression formed

491

(as noted by Hess 1998) by analogy with the productive relational use of -bid discussed in

Section 2.2.10 above:

Comparatives of equivalence are formed using the verb x&ʷul’ab ‘be that way, be similar to’:

(683) a. luƛ’ čəxʷ x%ʷul’ab ʔə ʔəca

luƛ’ čəxʷ x%ʷul’ab ʔə ʔəca

old 2SG.SUB similar PR I

‘you’re as old as I am’ (lit. ‘you are old, similar to me’)

(Hess & Hilbert 1976: II, 52)

b. ƛ’al’ bəqa ti yiq’us x%ʷul’ab ʔə ʔəca

ƛ’al’ bə=qa ti yiq’us–s x%ʷul’ab ʔə ʔəca

also ADD=many SPEC cedar.root.basket–3PO similar PR I

‘she also has as many cedar-root baskets as I have’

(lit. ‘her cedar-root baskets are also many, similar to me’)

(Hess & Hilbert 1976: II, 53)

c. dəbəxʷ əw’ə tiʔiɬ p’q’ac tiʔiɬ sx%qalikʷyitəbs x%ʷul’ab ʔə tiʔiɬ tusəshuys

dəb=əxʷ əw’ə tiʔiɬ p’q’ac tiʔiɬ s=x%q•alikʷ–yi–t–əb=s

instead=now PTCL DIST rotten.wood DIST NM=bound•bundle–DAT–ICS–PASS=3PO

x%ʷul’ab ʔə tiʔiɬ tu=s=ʔəs–huy=s

similar PR PROX PAST=NM=STAT–be.done=3PO

‘so, instead it is a rotten log that is wrapped up for him as [the elk] had been’

(Hess 1998: 87, line 256)

In these constructions, x&ʷul’ab is placed in series with the predicate expressing the basis for

comparison (luƛ’ ‘old’ in 683a), the standard of comparison being expressed by its oblique object

(introduced by the “empty” preposition ʔə). When the comparee is a first- or second-person, it is

express by a subject clitic, as in (683a), and when it is expressed by an NP, it immediately

follows the basis of comparison and precedes x&ʷul’ab, as in (683b). Unlike the comparison of

inequality, the comparison of equivalence does not seem to be limited to scalar predicates but

instead can use regular verbal expressions as a basis of comparison (683c).

8.9 Possession

Unlike English, Lushootseed does not have a specific verb expressing the notion of

possession; instead, there are a variety of ways of expressing possession, the most frequent of

492

which (and the most closely comparable to have in terms of its uses) consists of a clause

predicated on the verb ʔa ‘be there’ with possessed noun as a subject:

(684) a. ʔa ti dtalə

ʔa ti d–talə

be.there SPEC 1SG.PO–money

‘I have money’

(Hess 1998: 30)

b. ʔa tsiʔiɬ čəgʷas

ʔa tsiʔiɬ čəgʷas–s

be.there DIST:FEM wife–3PO

‘he has a wife’

(Hess 1998: 78, line 7)

c. ʔaʔ tiʔiɬ dsqəlalitut ɬulaʔ čəd gʷəl ɬuqʷšab tiʔəʔ swatixʷtəd

ʔaʔ tiʔiɬ d–sqəlalitut ɬu=laʔ čəd gʷəl ɬu=qʷšab

be.there DIST 1SG.PO–spirit.power IRR=localize 1SG.SUB SCONJ IRR=foggy

tiʔəʔ swatixʷtəd

PROX country

‘I have a spirit power that I will locate and the world will be foggy’

(Hilbert & Hess 1977: 28)

Such constructions would have a literal translation of ‘there is/exists Y’s X’, where Y represents

the possessor and X the possessed item. These expressions are negated through existential

negation (Section 8.6.2):

(685) a. xʷiʔ kʷi gʷədsxʷiʔxʷiʔ

xʷiʔ kʷi gʷə=d–sxʷiʔxʷiʔ

NEG REM SBJ=1SG.PO–game

‘I have no game’

(Hess 1998: 80, line 60)

b. xʷiʔ kʷi stab gʷubəq’əd čəd gʷəʔəɬəd

xʷiʔ kʷi stab gʷə=ʔu–bəq’–d čəd gʷə=ʔəɬəd

NEG REM what SBJ=PFV–be.in.mouth–ICS 1SG.SUB SBJ=feed.on

‘I have nothing to put in my mouth to eat’

(Hilbert & Hess 1977: 20)

The ʔa possessive construction seems to be used in a wide range of contexts to express

possession in the literal sense (that is, physical possession of an object), true ownership, and

493

certain types of inalienable possession such as social and kinship affinities (684b), possession of

a spirit power (684c), etc.

When referring to possession in the future,

170

a figurative expression based on ʔatxʷ ‘put

something there’ (lit. ‘cause something to be there’) is used:

(686) a. x%ʷul’əxʷ čəxʷ ɬuʔatxʷ tsiʔiɬ q’ədᶻax% ʔə tsiʔiɬ sqigʷac

x%ʷul’=əxʷ čəxʷ ɬu=ʔa–txʷ tsiʔiɬ q’ədᶻax% ʔə tsiʔiɬ sqigʷac

only=now 2SG.SUB IRR=be.there–ECS DIST:FEM intestines PR DIST:FEM deer

‘you will just have those intestines of that Deer’

[ML Mink and Tutyika I, line 210]

b. bək’ʷ čələp ɬuliɬʔatxʷ

bək’ʷ čələp ɬu=liɬ–ʔa–txʷ

all 2PL.SUB IRR=PRLV–be.there–ECS

‘you guys will have everything along there’

(Hess 2006: 66, line 598)

Unlike the ʔa possessive, the ʔatxʷ construction expresses the possessor as the subject of the

verb and the (to be) possessed item as its direct object.

Possession is also occasionally expressed through “loose” predicate nominals in which the

possessed item is realized as a predicate nominal and the possessor as a syntactic subject:

(687) wiˑʔ, x%ʷul’əxʷ čəd ɬusp’ic’ikʷ

wiʔ x%ʷul’=əxʷ čəd ɬu=sp’ic’ikʷ

declare only=now 1SG.SUB IRR=diaper.child

‘she declares, now I will have a diaper child’

[DS Star Child, line 159]

These constructions are extremely rare in the textual corpus (limited in fact to this single instance

of reported speech), but are much more frequent in conversation (Hess, p.c.). Given the

importance of context for the correct interpretation of such utterances, it seems likely that these

are informal, pragmatically-marked expressions.

Another, more specific, method of building possessive constructions is through the use of the

propriative prefix bəs- (Section 2.1.5):

170

There are, in fact, no uses of the ʔa possessive construction in any tense except the present in the current corpus

494

(688) a. ʔəbsbədaʔ ʔə tiʔəʔ stubš

ʔəs–bəs–bədaʔ ʔə tiʔəʔ stubš

STAT–PROP–offspring PR PROX man

‘they have a son’

b. tabsʔic’əb čəd ʔə ti x%ʷiqʷəq’ʷ

tu=ʔas–bəs–sʔic’əb čəd ʔə ti x%ʷiqʷəq’ʷ

PAST=STAT–PROP–blanket 1SG.SUB PR SPEC white

‘I have a white blanket’

(Hess 1998: 30)

This prefix is added to nouns meaning ‘X’ to form verbs meaning ‘have X’. Propriative verbs are

generally confined to the expressions involving kinship, social relations, and ownership.

Possessive constructions can also be formed using noun phrases containing the particle gʷəɬ

‘associative’:

(689) a. hay, gʷəɬəxʷ sqigʷac tiʔiɬ q’ədᶻax% ʔə tsiʔiɬ p’uay’

hay gʷəɬ=əxʷ sqigʷac tiʔiɬ q’ədᶻax% ʔə tsiʔiɬ p’uay’

SCONJ ASSC=now deer DIST intestines PR DIST flounder

‘so, Flounder’s intestines belong to Deer now’

[ML Mink and Tutyika I, line 249]

b. yəx%i huy gʷəɬ sqigʷac

yəx%i huy gʷəɬ sqigʷac

because SCONJ ASSC deer

‘because they were Deer’s’

[ML Mink and Tutyika I, line 192]

However, gʷəɬ ‘associative’ itself is not so much as expression of possession as it is an expressed

of some sort of associative relationship that can be used to identify a particular item or type of

item, one of the most salient relationships for such identification being possession. Noun phrases

containing gʷəɬ are discussed in more detail in Section 7.3.3

495

9 Complex sentences

9.1 Verb series

d. gʷəl x%ʷul’ čəd gʷət’əbaʔagʷil čəda gʷəʔux%ʷ, gʷəʔusil, gʷət’ičib

gʷəl x%ʷul’ čəd gʷə=t’əbaʔ–agʷil čəda gʷə=ʔux%ʷ

then only 1SG.SUB SBJ=fall.in.water–AUTO 1SG.COORD SBJ=go

gʷə=ʔusil gʷə=t’ičib

SBJ=dive SBJ=swim

‘then I would just jump in the water, I’d go and dive and swim [away]’

[ML, Mink and Tutyika, line 32]

9.2 Co-ordination

9.3 Subjunctive subordinate clauses and embedded interrogatives

d. ƛ’ub ʔa ʔal kʷi čad ɬucəxʷəsbəč, ɬudəxʷʔatubšləp tiʔəʔ sʔəɬəd

ƛ’ub ʔa ʔal kʷi čad ɬu=d=dəxʷ=ʔəs–bəč

well be.there at REM where IRR=1SG.PO=ADNM=STAT–fallen

ɬu=dəxʷ=ʔa–txʷ–bš=ləp tiʔəʔ sʔəɬəd

IRR=ADNM=be.there–ECS–1SG.OBJ=2PL.PO PROX food

‘there should be food wherever I am laid, wherever you guys will put me’

(Hess 1998: 92, line 31)

9.4 Sentential complements

haydx„ \lg„\˙ stubß ti˙i¬

know plural man det

she knew that he was a man.

til\b ˙uhaydx„ ˙ul\k’„t\b ti˙i¬ ˙alalß ˙\ tsi sx„iyuk’„

til\b ˙u–haydx„ ˙u–l\k’„–t–\b ti˙i¬ ˙al–alß ˙\ tsi sx„iyuk’„

suddenly pfv–know pfv–eat–ics–pass det pl–cross.sex.sibling Pr det:fem

Basket.Ogress

right away he found out that his silbings were eaten by the Basket Ogress.

JS Basket Ogress 19

9.5 Adverbial clauses

used as adverbial without determiner, Coyote and daughter line 273, 278, and

496

(690) a. ʔuhuyud sup’ayəqs

ʔu–huyu–d s=ʔu–p’ayəq=s

PFV–be.done–ICS NM=PFV–carve.canoe=3PO

‘he did his carving’

(Hess 2006: 47, line 125)

b. ʔəskʷukʷcut ʔə tiʔəʔ sušəɬs səsqʷalc

ʔəs–kʷukʷcut ʔə tiʔəʔ s=ʔu–šəɬ=s s=ʔəs–qʷalc=s

STAT–cook PR PROX NM=PFV–make=3PO NM=STAT–boil=3PO

‘he was cooking what he had put on to boil’

(Hess 1998: 94, line 91)

c. ti tusƛ’iqačiʔbtub čəɬ ɬuhuyutəb čəɬ sqadaʔs

171

ti tu=s=ƛ’iq•ačiʔ–b–txʷ–b čəɬ ɬu=huyu–t–əb čəɬ

SPEC PAST=NM=sticky•hand–MD–PASS 1PL.PO IRR=be.done–ICS–PASS 1PL.SUB

s=qadaʔ=s

NM=steal=3PO

‘our hands getting stuck would be what caused us to be stolen’

(Hess 2006: 71, line 706)

adverbial (OVERLAP WITH DEXW-)

cause/motive

tul\sc’ud\x„ k„i tu(s)sax„\bs

tu=l\s–c’ud–\x„ k„i tu=s–sax„\b–s

past=cont–weak–now det past=np–jump–3po

He had become weak from his running.

Little Diver line 118

≈„ul’ ç\x„ ˙uhuyud ˙\s˙ista˙ stab k„(i) adsd¸aÒ’ad

≈„ul’ ç\x„ ˙u–huyu–d ˙\s–˙ista˙ stab k„i ad–s–d¸aÒ’a–d

only 2sg.sub pfv–be.done–ics stat–be.like what det 2sg.po–np–

confused–ics

You just make it like that [so] you confuse him.

Coyote’s son line 32

location

Ò’usug„adg„ads k„i Ò’us˙alils k„i d\x„g„adg„ads

Ò’u–s–˙u–g„adg„ad–s k„i Ò’u–s–˙alil–s k„i d\x„–g„adg„ad–s

hab–np–pfv–speak–3po det hab–np–get.to.place–3po det np2–

speak–3po

she is there chattering when she converses.

Owl line 184

171

The final verbs in this sentence is recorded in the original text with the possessive subject-maker =s; however,

this marker would normally be required by the nominalization of the verb.

497

˙\bil’ ç\x„ b\t’aq’t ˙al [ti] swatix„t\d k„i ¬(u)adsud¸\k’„

˙\bil’ ç\x„ b\–t’aq’t ˙al ti swatix„t\d k„i ¬u–ad–s–˙u–d¸\k’„

cond 2sg.sub add–ahsore Pr det country det irr–2sg.po–np–pfv–

wander

Perhaps you will go up into the high country too [where] you will wander around

Changer 240

temporal

¬u¬a≈ k„i [s]cut[s], [k„i s]Ò’\ladi˙s

¬u–¬a≈ k„i s–cut–s k„i s–Ò’\ladi˙–s

irr–dark det np–say–3po det np–make.noise–3po

It will be night when he talks, when he makes noise.

no det

huy g„\l dx„caq’a≈adid\x„ put (s)ß\qla≈ad\bs ˙\ ti˙i¬ c\di¬ ç’\t≈

huy g„\l dx„–caq’–a≈ad–di–d–\x„ put s–ß\qla≈ad\b–s ˙\ ti˙i¬ c\di¬

ç’\t≈

intj conj ctd–spear–side–ss–ics–now really np–raise.arms–3po Pr det he

kingfisher

And he speared him in the side just as he raised his arms (wings)

little Diver line 91 (NO DET)

WITH PREPOSITIONS (DEXW ALSO USED THIS WAY)

temporal clause

g„\l l\k’„\dax„ ˙\ ti˙i¬ suk„iç’yit\bs

g„\l l\k’„\d–ax„ ˙\ ti˙i¬ s–u–k„iç’–yi–t–\b–s

conj eat–now Pr det np–pfv–butcher–ben–ics–md–3po

‘And he ate it as it was butchered for him.’

≈„ul’ l\cu˙i˙¬ad\b ˙\ ti˙\˙ sq’„\la¬\d ˙\ ti˙i¬ Ò’usq’„\ls

≈„ul’ l\cu–˙i–˙¬ad–\b ˙\ ti˙\˙ sq’„\la¬\d ˙\ ti˙i¬ Ò’u–s–q’„\l–s

only cont–dim–eat–pass Pr det berry Pr det hab–np–ripe–3po

He was [simply] eating berries [directly off the bush] as they ripened.

Bear and Ant, line 20

no det

x„i˙\x„ g„\b\s≈„ubils dx„˙al\x„ sbi¬a˙il\x„ ˙\ ti˙\˙ t\k„t\k„\lus

x„i˙–\x„ g„\–b\–s–≈„ubil–s dx„˙al–\x„ s–bi¬a˙il–\x„ ˙\ ti˙\˙ t\k„t\k„\lus

neg–now sbj–add–np–be.quiet–3po towards–now np–be.fed.up–

now Pr det owl

She [would] not keep quiet until [finally] Owl got fed up.

498

d. gʷəl hikʷ ʔuhiiɬ əlgʷəʔ ʔə tiʔəʔ shəliʔdubs əlgʷəʔ ʔə tiʔiɬ sxʷəctəbs tiʔiɬ

dəxʷuʔatəbəds əlgʷəʔ

gʷəl hikʷ ʔu–hiiɬ əlgʷəʔ ʔə tiʔəʔ s–həliʔ–dxʷ–b–s əlgʷəʔ ʔə tiʔiɬ s–

xʷəc–t–əb–s tiʔiɬ dəxʷ–ʔu–ʔatəbəd–s əlgʷəʔ

conj big pfv–happy plural Pr det nm–alive–lc–pass–3po plural Pr

det nm–extract–ics–pass–3po det np2–pfv–die–3po plural

And they were very glad for having been saved by the removal [of the quills] which had been

why they had died. SH 491 + 64

(691) baˑlgʷas buʔqʷ kʷi bək’ʷ sčads ʔə ti buʔqʷ gʷəɬ x%ʷəlč, tiʔiɬ bəgʷəɬ t’aq’t, gʷəɬ spaɬx%ad

buʔqʷ

balgʷas buʔqʷ kʷi bək’ʷ s=čad=s ʔə ti buʔqʷ

all.kinds waterfowl REM all NM=where=3PO PR SPEC

waterfowl

gʷəɬ x%ʷəlč tiʔiɬ bə=gʷəɬ t’aq’t gʷəɬ

ASSC sea DIST ADD=ASSC inland ASSC

spaɬx%ad buʔqʷ

tidal.flats waterfowl

‘there were all kinds of Duck People (waterfowl) that came from everywhere, Duck

People from the sea and those from inland, Duck People from the tidal flats’

(Hess 2006: 63, line 457)

499

10 Locative and spatial expressions

10.1 Prepositional phrases

10.2 Locative and directional adverbials

10.3 Demonstrative adverbs

500

11 Communicative structure

Communicative Structure (Mel’čuk 2001), also known as Information Structure (Lambrecht

1994) or Information Packaging (Foley and van Valin 1985), may be loosely defined as those

aspects of sentence structure that are conditioned by the communicative intent of speakers (as

opposed to the semantic content of the utterance) and/or by the context in which the sentence is

spoken. Although it is often left aside in grammatical descriptions of natural languages as being

outside the domain of morphosyntax, Communicative Structure in fact has a profound influence

on many aspects of Lushootseed sentence structure, including processes such as the selection of

sentence predicates, the ordering of constituents, and the direct expression of communicative

properties of utterances by lexical or syntactic means. The following sections will discuss those

aspects of Lushootseed morphosyntax that are most relevant to the encoding of the

Communicative Structure of sentences, concentrating in particular on two categories that pertain

to Communicative Structure per se — Thematicity and Focalization — and a third aspect of the

organization of discourse which is not proper to but rather derived from Communicative

Structure — (discourse) Topicality.

172

Of these three, the communicative category of Thematicity is probably the most familiar,

having been promoted by, among others, Halliday (1970) and being the equivalent of the widely

recognized division of the sentence into Topic and Comment proposed by Sapir (2004).

Thematicity entails the division of most sentences into two portions — the Theme, which is

(loosely speaking) what the sentence is “about,” and the Rheme, which is what is being said

“about” the Theme (Mel’čuk 2001: 95ff). Focalization (not to be confused with the term

172

Because the literature in this area is quite diverse and there is a lack of consensus in the field on a widely-

accepted and well-defined terminology for its analysis, for the purposes of the discussion below I have adopted the

definitions of terms proposed by Mel’čuk (2001). Although these definitions are presented somewhat informally in

what follows, readers are referred to this work for detailed examination of the intricacies of the concepts and the

technical details of their implementation in a comprehensive model of Communicative Structure, which necessarily

includes many more considerations that can be dealt with within the limitations of the present work.

501

“Focus,” which as used by Lambrecht 1994 and others is a synonym for Rheme), on the other

hand, refers to the designation of an event-participant or some aspect of an event (such as a

place, time, or setting) as being “logically prominent” or the focal point of a speaker’s attention

(Mel’čuk 2001: 181). Frequently, though not always, a Focalized element is in some way

contrastive or in opposition to some other element (i.e., whatever is being said of the Focalized

element is also being said to be explicitly not true of some other element of the discourse).

Thematicity and Focalization are independent of one another, and a Focalized element may be a

Theme, a Rheme, or some part of either one.

Both Thematicity and Focalization are sentence-level categories — that is, they are defined

within the bounds of a single sentence. In contrast, a discourse Topic is defined over multiple

sentences and/or entire discourse episodes: specifically, a discourse Topic is a Theme which is

shared by a large number of sentences in sequence or within the bounds of a delimitable stretch

of connected discourse (cf. Mel’čuk 2001: 209). This use of the term is different from the

common use of “Topic” as (essentially) a synonym for Theme — indeed, the fact that discourse

Topics are necessarily Themes at the sentence-level has led to a certain degree of confusion, as

well as the common misapprehension that the two (Theme and discourse Topic) are the same.

However, if Topics are necessarily Themes, it is not true that Themes are necessarily discourse

Topics, and given that the definitions employed here define the Theme at sentence level and the

Topic at the level discourse, it is important to keep the two distinct. This also means that, strictly

speaking, discourse Topics are not elements of the Communicative Structure of individual

sentence, but belong instead to the realm of text-planning; nevertheless, as we will see in Section

11.3 below, the notion of discourse Topic is crucial to the interpretation of Lushootseed

utterances in context, and so has a major impact on sentence-level morphosyntax.

502

11.1 Theme and Rheme

The aspect of Communicative Structure that has the most profound influence on syntax is

Thematicity — the division of the sentence between Theme and Rheme. The primary effect of

Thematicity on syntactic structure manifests itself in a near-absolute constraint that clausal

subject, especially third-person subjects, be Thematic and that sentence predicates be

Rhematic.

173

This constraint manifests itself in a number of ways, both at the level of discourse

and the contextual interpretation of sentences (see Section 11.3) and at the most basic level of

clausal syntax, where it governs the selection of syntactic predicate. Where most languages

select the lexical items used as sentence-predicates based largely on their part of speech (that is,

they choose sentence predicates that are verbs), Lushootsed selects that element of the sentence

that is Rhematic to be sentence predicate, irrespective of its part of speech. This pattern is

illustrated by the contrast between the question-and-answer pairs in (692):

174

(692) a. i. ʔuʔəx#id kʷi kikəwič

ʔu–ʔəx%id kʷi ki–kəwič

PFV–what.happen REM ATTN–hunchback

‘what happened to Little Hunchback?’

ii. ʔuˑ, ƛ’al’ busaxʷəbdubut tiʔiɬ kikəwič

ʔu ƛ’al’ bə=ʔu–saxʷəb–dxʷ–but tiʔiɬ ki–kəwič

INTJ also ADD=PFV–run–DC–REFL DIST ATTN–hunchback

‘oh, Little Hunchback also managed to escape’

[DM Basket Ogress, lines 79 – 80]

b. i. stab ti ʔučalatəb ʔə tiʔiɬ wiw’su

stab ti ʔu–čala–t–əb ʔə tiʔiɬ wiw’su

what SPEC PFV–hit–ICS–PASS PR DIST children

‘what was chased by the children?’

173

Cf. Beck (2000b), where this is discussed using the term “Topic” instead of Theme. Kinkade (1990) makes the

same observation for Salishan in general, also using the term “Topic.” As noted in Section 11.3, the alignment of

subject and Theme at the clause level coincides with an alignment of subject and Topic at the discourse level,

resulting in a three-way alignment of Theme, subject, and discourse Topic.

174

Question-and-answer pairs such as these are frequently-touted diagnostics for (or even definitions of) the

Thematic divisions of sentence, the Theme being equated with the presupposed portion of the question (and its

repetition in the answer) and the Rheme with the new information contained in the answer (and the interrogative

element in the question).

503

ii. sqʷəbayʔ ti ʔučalatəb ʔə tiʔiɬ wiw’su

sqʷəbayʔ ti ʔu–čala–t–əb ʔə tiʔiɬ wiw’su

dog SPEC PFV–hit–ICS–PASS PR DIST children

‘the one chased by the children is a dog’

(Hess 1995: 98 – 99)

The question in (692a-i) is a narratively-focused question asking about an event in which a

particular Thematic event-participant, kikəwič ‘Little Hunchback’, is involved, and elicits a

narratively-focused response with a Rhematic verbal predicate (692a-ii), recounting what it is

that Little Hunchback did. The question in (692b-i), however, is an information question (Section

8.4.2) and asks for the identity of an unknown participant in a Thematic event (the children’s

chasing), this event being expressed as a headless relative clause (ti ʔučalatəb ʔə tiʔiɬ wiw’su

‘what was chased by the children’) in subject position of a sentence whose predicate is the

interrogative word stab ‘what?’. The response in (692b-ii) mirrors this structure exactly,

substituting the requested information (the identity of the chased even-participant, sqʷəbayʔ

‘dog’) for the interrogative word, giving us a sentence with a Rhematic nominal predicate of the

type discussed in Section 8.3.1.

This constraint governing nominal and other types of non-verbal predication also holds

outside of the context of questions, occurring in narrative and other discourse contexts, where the

event itself is Thematic (and, generally, Given) and the Rheme is an event-participant or some

other non-verbal element of the sentence, as in (693). This sentence occurs at a point in a story

where the speakers, a pair of supernatural hunters, have rewarded the protagonist, Pheasant, for

his modesty by giving him game, which they have packed and magically made light so that

Pheasant can carry it. As they finished preparing the game, the hunters explain to Pheasant what

they have done, and remind him why making the pack light was important:

(693) hikʷ kʷagʷičəd tiʔiɬ səsɬild čəɬ ti dəgʷi

hikʷ kʷagʷičəd tiʔiɬ s=ʔəs–ɬil–d čəɬ ti dəgʷi

big elk DIST NM=STAT–give.food–ICS 1PL.SUB SPEC you

‘what we haven given you is a big elk’

(Hess 1998: 81, line 110)

504

In (693), what is Rhematic is the nature of the gift, while the fact that the hunters are giving

something to Pheasant is old news and entirely Thematic. As a result, the sentence predicate is

the noun phrase hikʷ kʷagʷičəd ‘big elk’ and the syntactic subject is a sentential nominal formed

on the expression of the event tiʔiɬ səsɬild čəɬ ti dəgʷi ‘what we have given to you’. In spoken

English, a sentence like this would probably be marked by prosody (we are giving you a big

elk), and in writing such sentences tend to be translated as clefts.

The requirement that the Rhematic element of a sentence be the syntactic predicate thus has

profound influence on the structure of Lushootseed sentences, and — in combination with the

flexibility which Lushootseed shows with respect to which parts of speech are eligible syntactic

predicates — is responsible for many of the exotic-looking structures with non-verbal predicates

discussed in more detail in Section 8.3 above. In (694), for example, the Thematic organization

of the sentence results in a complex structure with a sentential nominal as its syntactic predicate:

(694) dsəsʔabyitəb ʔə tiʔiɬ dsqa tiʔəʔ diʔəʔ cəxʷx%ix%ilix%txʷ tiʔəʔ diʔəʔ stawixʷəʔɬ

d=s=ʔəs–ʔab–yi–t–b ʔə tiʔiɬ d–sqa

1SG.PO=NM=STAT–extend–DAT–ICS–PASS PR DIST 1SG.PO–older.brother

tiʔəʔ diʔəʔ d=dəxʷ=x%i–x%ilix%–txʷ tiʔəʔ diʔəʔ stawixʷəʔɬ

PROX he 1SG.PO=ADNM=ATTN–fight–ECS PROX he children

‘this is what my older brother has given me so I can compete with the children’

[MW Star Child, line 100]

The context of the sentence is a situation where a character is explaining the nature of a

particular item (tallow), expressed in the sentence as the subject of the matrix clause, tiʔəʔ diʔəʔ

‘this here’. Under these circumstances, the object being discussed (the tallow) is Topical and

Thematic, whereas the Rheme is the speaker’s explanation of it — specifically, the fact that it

was given to him by his older brother. The act of giving is expressed as a sentential

nominalization (dsəsʔabyitəb ʔə tiʔiɬ dsqa, literally ‘my being given-to by my brother’) which is

used as a syntactic predicate. The nominalization of ʔabyitəb ‘be given something by somebody’

505

is required in that the gift, tiʔəʔ diʔəʔ ‘this here’, corresponds to an oblique object of the finite

verb (Section 7.4.2).

The same principles lie behind the selection of other types of non-verbal predicate. The

sentences in (695), for example, come from a point in a story where two brothers, Diaper Child

and Moon, make (as part of the process of creating the world) some moccasins. The narrator

goes on to explain how many they made, and how many belong to each brother:

(695) a. ʔəsbuus kʷi tuhuyud əlgʷəʔ

ʔəs–buus kʷi tu=huyu–d əlgʷəʔ

STAT–four REM PAST=be.made–ICS PL

‘it is four (moccasins) that they made’

b. saliʔ kʷi gʷəɬ sp’ic’ikʷ

saliʔ kʷi gʷəɬ sp’ic’ikʷ

two REM ASSC Diaper.Child

‘those belonging to Diaper Child [are] two’

c. saliʔ kʷi gʷəɬ sɬukʷalb

saliʔ kʷi gʷəɬ sɬukʷalb

two REM ASSC moon

‘those belonging to Moon [are] two’

[DS Star Child, lines 355 – 357]

The predicates of these sentences are numerals — buus ‘four’ in (695a) and saliʔ ‘two’ in (695b)

and (c). These numbers constitute the Rhematic portion of all of these sentences, the essential

element of the communication being the number of moccasins (rather than the event of making

the moccasins or the fact that it was moccasins that were made), and so they are used as the

syntactic predicate.

Precisely the same conditions governing the selection of clausal predicates allows for the

predicative use of other phrases-types, such as the prepositional phrase heading the clause in

(696):

506

(696) ʔaləxʷ sbuusaɬdatils kʷi suq’ilaʔkʷčups sp’ip’ic’ikʷ

ʔal=əxʷ s=buus•aɬ•dat–il=s Ø kʷi s=ʔu–q’il•aʔkʷ•čup=s

at=now NM=four•times•day–INCH=3PO 3SUB REM NM=PFV–aboard•group•fire=3PO

sp’i–p’ic’ikʷ

ATTN–Diaper.Child

‘it was (i.e., happened) on the fourth day as Sp’ip’ic’ikʷ was loading firewood’

[HM Star Child, line 102]

This sentence occurs at a point in the narrative where the story-teller has recounted the various

tasks that the protagonist, who has been enslaved by Raven, is forced to carry out, one of which

is loading firewood onto Raven’s canoe. The line in (696) sets the scene for the subsequent

discourse episode by specifying the time at which the events occurred, the Rhematic portion of

the sentence being the temporal expression (the PP). As a result, the prepositional phrase ʔal

sbuusaɬdatils ‘on the fourth day’ is the syntactic predicate.

The close alignment between Rhematicity and expression as syntactic predicate can also be

seen to govern the syntax of lexical adverbs (Section 2.5.1), which can be used either ad-verbally

as predicate modifiers (697a) or as syntactic predicates in their own right (697b):

(697) a. ƛ’ub čəd ʔəsʔistaʔb ʔə tiʔəʔ

ƛ’ub čəd ʔəs–ʔistaʔb ʔə tiʔəʔ

well 1SG.SUB STAT–same–MD PR PROX

‘I had better (stay) like this’

(Hess 1998: 82, line 117)

b. ƛ’ub čəxʷ ʔə ti adsʔux%ʷ sgʷəlub

ƛ’ub čəxʷ ʔə ti ad=s=ʔux%ʷ sgʷəlub

well 2SG.SUB PR SPEC 2SG.PO=NM=go Pheasant

‘you will be all right as you go, Pheasant’

(Hess 1998: 81, line 101)

The first sentence, which is “about” the speaker (making the first-person Thematic), tells the

addressee what the speaker had better do, making the action and its qualification Rhematic. As

part of the Rheme, ƛ’ub ‘well’, is included as part of the syntactic predicate — specifically, as an

adverbial modifier of ʔistaʔb ‘be the same as something’. In the next example, however, the

507

assertion is that the addressee (the second person) will be well, making ƛ’ub Rhematic and

requiring that it be syntactic predicate, while the Thematic second-person is expressed as subject.

Thematicity also underlies the structure of the existential negative (Section 8.5), which is

used to negate the existence of something (698a) or its presence at a specific location (698b):

(698) a. xʷiʔ kʷi gʷat gʷəhəliʔ

xʷiʔ kʷi gʷat gʷə=həliʔ

NEG REM who SBJ=alive

‘no one could survive’

(Hess 2006: 79, line 889)

b. xʷiʔəxʷ tiʔəʔ tubədaʔs

xʷiʔ=əxʷ tiʔəʔ tu=bədaʔ–s

NEG=now PROX PAST=offspring–3PO

‘her child was gone’

[MW Star Child, line 121]

In each of these cases, the sentences are “about” the Thematic subjects and the communicative

import of the sentence is the negation. Thus, (698a) is a statement about potential survivors and

asserts that there are none (i.e., that they don’t exist); (698b) is about a child and tells the

addressee that the child is no longer there. In such cases, the negative adverb xʷiʔ is Rhematic

and is the syntactic predicate, while the Thematic elements are subjects. The choice between the

use of the existential negative construction and adverbial negation (discussed in Section 8.5

above) is governed by the same principles that govern the uses of adverbs illustrated in (697).

Because Lushootseed is, generally speaking, a predicate-initial language, the requirement

that syntactic predicates be Rhematic means that Lushootseed is also effectively a Rheme-initial

language. Another effect of Thematicity on constituent order can be seen in passive constructions

(Section 6.2) which — rather unusually for Lushootseed — allow alternative orderings of their

arguments, showing either agentive complement (AgCo) >> subject (Sub) order, as in (699a), or

Sub >> AgCo order, as in (699b):

508

(699) a. ʔu–gʷəč’–t–b ʔə ti č’ač’as ti sqʷəbayʔ

PFV–seek–ICS–PASS PR SPEC child SPEC dog

‘the dog was looked for by the boy’

b. ʔu–gʷəč’–t–b ti sqʷəbayʔ ʔə ti č’ač’as

PFV–seek–ICS–PASS SPEC dog PR SPEC child

‘the dog was looked for by the boy’

(Hess 1995: 23, exx. 6a – 6b)

Although the expected order, following the general principle (for NP arguments) that the subject

precedes other arguments of the verb, would be that in (699b), in elicitation (Hess, p.c.) and in

texts the more common order is that shown in (699a).

175

Examination of the contexts in which

the two orders occur suggests that the choice between the two is governed to a large extent by

Thematicity, and falls out from the preference for Rheme-initial sentences, the AgCo >> Sub

order occurring most frequently in situations where both the agentive complement and the verb

are Rhematic, thereby maintaining the linear contiguity of the Rheme. This preference for

contiguity of the Rheme outweighs the preference for the subject NP to immediately follow the

matrix clause predicate that governs constituent-order in other sentence types, resulting in a non-

canonical ordering of syntactic arguments.

The greater frequency of the AgCo >> Sub order in Lushootseed passives stems from the

strong Lushootseed preference for Topical (and, hence, Thematic) subjects (see Section 11.3

below). Although Topical NPs are most frequently elided in Lushootseed discourse, there are

certain contexts where it is necessary to make the Topic of discourse explicit. The most frequent

of these in texts is a topic-setting construction like that shown in (700):

(700) kʷukʷcutyitəbəxʷ ʔə tiʔəʔ c’ix%c’ix% tiʔiɬ syaʔyaʔs, sčətxʷəd

kʷukʷcut–yi–t–b=əxʷ ʔə tiʔəʔ c’ix%c’ix% tiʔiɬ syaʔyaʔ–s sčətxʷəd

cook–DAT–ICS–PASS=now PR PROX fish.hawk DIST friend–3PO bear

‘his

friend Bear was cooked for by Fish Hawk

’

(Hess 1995: 154, line 68)

175

A rough count of matrix-clause passives with both an NP subject and an agentive complement phrase in the

current corpus shows 14 examples of Sub >> AgCo order as opposed to 37 examples of AgCo >> Sub order.

509

This sentence occurs at the beginning of a discourse episode where the narrator is shifting Topic

by making a statement “about” (Fish Hawk’s friend) Bear, asserting that he has had some food (a

salmon caught in the previous episode) prepared for him by Fish Hawk. Bear, as the intended

new discourse Topic for the following episode, is Thematic, and the Rhematic portion of this

sentence includes both the event of food preparation and the fact that it was Fish Hawk who

prepared the food. The order of constituents here thus ensures the contiguity of the Rheme, in

spite of the fact that it violates the usual predicate >> subject pattern.

The alternative constituent-order, Sub >> AgCo, occurs typically in those cases where the

entire sentence, including both the subject and the agentive complement, is Rhematic. In

narratives, such sentences occur in contexts where an event involving participants that are not

Topical at that point in discourse is introduced, usually at the beginning of a new discourse

episode. In the excerpt in (701), for instance, the passive clause in (701d) introduces a new event

involving two event-participants, the Dwarves and the Duck People, who are not mentioned in

the previous three lines (which are concerned with the protagonists of the story, the Seal Hunters,

who have been captured by the Dwarves):

(701) a. huy, xʷuʔəʔəʔ xʷiʔ ləhaʔkʷ kʷi tushədʔiw’s əlgʷəʔ

huy xʷuʔəʔəʔ xʷiʔ lə=haʔkʷ kʷi tu=s=hədʔiw’=s əlgʷəʔ

SCONJ maybe NEG NEGP=long.time REM PAST=NM=be.inside=3PO PL

‘then, I guess, they [the seal hunters] had not been inside very long’

b. gʷəl huy, huy, gʷəcutadəxʷ tiʔiɬ tul’šəq

gʷəl huy huy gʷəcutad=əxʷ tiʔiɬ tul’–šəq

SCONJ SCONJ SCONJ be.commotion=now PROX CNTRPT–be.high

‘and then, then there was noise from above’

c. gʷəcutadəxʷ kʷədiʔ tul’šəq

gʷəcutad=əxʷ tiʔiɬ tul’–šəq

be.commotion=now PROX CNTRPT–be.high

‘there was noise from above’

510

d. gʷəhaw’əʔ ʔušidᶻtəbəxʷ tiʔəʔ caadiɬ qʷiqʷqʷistay’bixʷ ʔə tiʔəʔ buʔqʷ

gʷəhaw’əʔ ʔu–šidᶻ–t–b=əxʷ tiʔəʔ caadiɬ qʷi–qʷ–qʷistay’bixʷ

seemingly PFV–be.surprised–ICS–PASS=now PROX they ATTN–ATTN–dwarves

ʔə tiʔəʔ buʔqʷ

PR PROX waterfowl

‘it seems that these dwarves were surprised by a sneak attack by the Duck People’

(Hess 2006: 61, lines 453 – 456)

The sentence in (701d) has only a Rhematic portion and would be a felicitous answer to a

question like “what happened next?,” in which none of the elements of the answer are

presupposed. Sentences such as this are frequently found at the beginning of discourse episodes

and at points in narrative where the narrator’s focus shifts between the actions of characters.

Because this type of sentence lacks a Theme, there is no conflict between the requirement that

the Rheme be linearly contiguous and the preference for predicate >> subject constituent order,

giving us the Sub >> AgCo order seen in (701d).

The same constituent order is also found in another situation in which the subject and

agentive completive belong to the same Thematic division of the sentence — in cases where both

are part of the Theme, as in (702):

(702) hay, c’əldub tiʔiɬ sčətxʷəd ʔə tiʔiɬ c’ix%c’ix%

hay c’əl–dxʷ–b tiʔiɬ sčətxʷəd ʔə tiʔiɬ c’ix%c’ix

SCONJ defeated–DC–PASS PROX bear PR PROX fish.hawk

‘and so Bear was defeated by Fish Hawk’

(Hess 1995: 154, line 93)

With this sentence, found towards the end of Edward Sam’s “tiʔiɬ sčətxʷəd ʔə tiʔiɬ c’ix&c’ix”

‘Black Bear and Fish Hawk’, the narrator begins a wrap-up and summary of the story. In this

particular context, both the subject, sčətxʷəd ‘Black Bear’ and the agentive complement, c’ix&c’ix

‘Fish Hawk’, are Thematic, and the import of the sentence is to relate what happened to these

two protagonists who have been well-established in discourse (the sentence in essence answering

the question “what happened to Black Bear and Fish Hawk?”). Thus, (702) conforms to the

general constituent-ordering constraint that the subject immediately follow the matrix predicate,

511

which expresses the Rhematic part of the utterance (that is, the narrator here is re-capping

events), while maintaining the linear contiguity of the Rheme. As with the selection of the

syntactic predicate of sentences discussed earlier, the alternation of constituent ordering in

passives provides a good illustration of the centrality of Thematicity in the construction of

Lushootseed sentences.

11.2 Focalization

Although the effects of Focalization on the structure of Lushootseed sentences are perhaps

not as profound as are those of Thematicity, Focalization is an important component of

Lushootseed grammar, both at the level of morphosyntax and at the level of text-planning and

the organization of discourse. The discussion here will centre on two different means of

expressing Focalization — the focalizing adverb diɬ (Section 11.2.1) and a construction

involving the sentential conjunction gʷəl and the left-dislocation of the clausal subject (11.2.2).

11.2.1 Focalization with diɬ ‘just that one’

The most textually frequent means of implementing Focalization is through the use of the

adverb diɬ ‘just that one’, which both focalizes a Thematic subject and particularizes the referent

of that subject as the only entity in current discourse space to which the predicate of the sentence

applies. Syntactically, diɬ belongs to the class of lexical adverbs (Section 2.5.1) and so may itself

be the main predicate of a clause, as shown in (703):

(703) a. diɬ tiʔiɬ

diɬ tiʔiɬ

FOC DIST

‘that is just the one’

(Hess 1998: 87, line 259)

512

b. diɬiɬ əw’ə tiʔəʔ saʔ luƛ’

diɬ–iɬ əw’ə tiʔəʔ saʔ luƛ’

INTNS–just.that.one PTCL PROX bad old

‘that very bad old man is just the one’

(Hess 2006: 31, line 233)

c. diɬ tiʔəʔ ʔuɬčisəbš

diɬ tiʔəʔ ʔu–ɬčil–s–bš

FOC PROX PFV–arrive–ALTV–1SG.OBJ

‘the one who came to me is just that guy’

(Hess 1998: 99, line 220)

(703a) shows diɬ as the predicate of a simple clause whose subject is the demonstrative tiʔiɬ. In

(703b), diɬ — in Type III reduplicated form, indicating intensivity (Section 5.3.2) — is modified

by the predicate particle əw’ə, which in the absence of modifying adverbs or adverbial particles,

immediately follows the sentence predicate (Section 2.5.3). The sentence in (703c) shows diɬ

acting as a predicate and taking a headless relative clause as its subject. When used in this way as

the predicate of a sentence, diɬ indicates that some previous element of discourse — either the

discourse Topic (704a) or the preceding set of events (704b) — corresponds to its subject:

(704) a. diɬ tiʔəʔ sč’istxʷs

diɬ tiʔəʔ sč’istxʷ–s

just.that.one PROX husband–3PO

‘her husband was him (Coyote’s son)’

(Hess 2006: 33, line 288)

b. ʔuˑ, diɬəxʷ tiʔəʔ dəxʷʔəsčcilsəxʷ

ʔu diɬ=əxʷ tiʔəʔ dəxʷ=ʔəs–čc–il=s=əxʷ

INTJ just.that.one=now PROX ADNM=STAT–red–INCH=3PO=now

‘oh, why he (Kingfisher) is red now is (because of) that’

(Hess 2006: 15, line 102)

The subject of (704a) is the NP tiʔəʔ sč’istxʷs ‘her husband’, which is linked by diɬ to the

Topical event-participant (in this context, Coyote’s son, who is recognized by Pigeon as her true

husband, as opposed to Coyote, who has been impersonating his son in order to steal his wives).

In (704b), the subject is an adjunctive nominal, dəxʷʔəsčcils ‘his reason for having become red’,

which is linked back to the preceding events in the narrative (specifically, that Kingfisher has

been speared under the arm by Heron in retaliation for sleeping with Heron’s wife).

513

As well as linking the subject to previous discourse and focalizing it, diɬ also attributes a

certain uniqueness or particularity to the subject’s referent, indicating that the referent of the

subject and only that referent corresponds to a particular event-participant or sequence of events,

as in (704):

(705) a. diɬ k’ʷəɬ kʷədiʔ tubšədəd lədxʷʔal tsiʔəʔ bədaʔ ʔə tiʔəʔ tusbiaw ʔuʔatəbəd

diɬ k’ʷəɬ kʷədiʔ tubšədəd lə=dxʷ–ʔal tsiʔəʔ bədaʔ ʔə

FOC QTV REM.DMA Sahaptin PROG=CNTRFG–at PROX:FEM offspring PR

tiʔəʔ tusbiaw ʔu–ʔatəbəd

PROX coyote PFV–die

‘he is it seems the warrior coming for the daughter of the late Coyote who died’

b. gʷəl diɬ tiʔəʔ tučaʔkʷ

gʷəl diɬ Ø tiʔəʔ tu=čaʔkʷ

SCONJ FOC 3SUB PROX PAST=seaward

‘he it is who has come to the coast’

(Hess 1998: 97, line 181 – 182)

The context for (704) is a point in a narrative where Coyote, who has faked his death as part of a

ruse to marry his own daughter, returns to his village disguised as a Sahaptin warrior. Coyote had

told the people beforehand that, if he died, his daughter was to be married to a warrior from

inland, and in (704a), one of the characters declares that the stranger who has appeared among

them must be the man foreseen in Coyote’s prediction, then repeats in (704b) that this man from

inland, and no other, must be the very one referred to by Coyote.

When used as an adverbial modifier of a verbal predicate, diɬ has essentially the same

function. As expected of an adverb, diɬ appears in such cases in pre-verbal position, as shown in

(706a), and attracts both sentence-second predicate particles (706b) and matrix subject-clitics

(706c):

514

(706) a. diɬ ƛ’utux%ʷalikʷ dxʷgʷəd

diɬ ƛ’u=tux%ʷ–alikʷ Ø dxʷ–gʷəd

FOC HAB=stretched–ACT 3SUB CNTRPT–down

‘he’s the one who stretches things downwards’

176

(Hess 2006: 28, line 151)

b. diɬ əw’ə higʷəxʷ ʔudxʷqədidəxʷ tiʔəʔ sʔušəbabdxʷ sbəq’ʷaʔ ʔə tsiʔəʔ čəgʷas,

x%ʷuʔx%ʷəyʔ

diɬ əw’ə hikʷ=əxʷ ʔu–dxʷ–qəd–i–d=əxʷ Ø tiʔəʔ sʔušəbabdxʷ

FOC PTCL big=now PFV–CTD–fornicate–SS–ICS=now 3SUB PROX poor.guy

sbəq’ʷaʔ ʔə tsiʔəʔ čəgʷas–s x%ʷuʔx%ʷəyʔ

Heron PR PROX:FEM wife–3PO helldiver

‘indeed, he (Kingfisher) is the one who cuckolded poor Heron

with his

wife, Hell–

diver’

(Hess 2006: 14, line 77)

c. diɬ čəd ʔəsbaliicəxʷ tiʔiɬ

diɬ čəd ʔəs–bali–c=əxʷ tiʔiɬ

FOC 1SG.SUB STAT–forget–ALTV=now DIST

‘I forget which one it was (that found the monster)’

[ML Basket Ogress, line 220]

In (706a), the adverb modifies the verbal predicate tux&ʷalikʷ ‘stretch things’, focalizing the

Thematic subject (in this case, a zero third-person whose referent is Spider) and indicating that

the referent of this subject is a particular, and unique, individual in the context in which the

sentence is uttered. Likewise, in (706b) diɬ singles out and focalizes a particular individual

(Kingfisher) playing a very specific role in the narrative context (the story of Heron’s wife

Helldiver and her adulterous affair with Kingfisher), while (706c) focalizes the forgotten

individual, expressed as the distal demonstrative tiʔiɬ, and singles it out from amongst a

particular set of individuals (those characters in the story who might have found the monster

qiʔqəl’adiʔ). Similarly, the line in (707) comes from a point in a narrative where Mink, whose

176

The gloss of this sentence as a cleft (taken from the published source) arises from the communicative function of

diɬ as a marker of Focalization. As Focalization is frequently indicated in English through the use of clefts (Mel’čuk

2001), sentences with diɬ are often translated as clefts, although this is not an accurate reflection of their underlying

structure. Likewise, the translations of (706b) and (c) reflect the communicative rather than the syntactic structure of

the Lushootseed sentences.

515

roast salmon was stolen in the previous discourse episode, hears that Changer is traveling in the

area and realizes that Changer must be the thief, leading Mink to exclaim:

(707) diɬ ʔuqadadic

diɬ ʔu–qada–di–t–s Ø

FOC PFV–steal–SS–ICS–1SG.OBJ 3SUB

‘he’s the one who robbed me of it’

(Hess 1998: 67, line 67)

Here, the adverb diɬ focalizes the Thematic subject, the Ø third-person pronominal whose

antecedent is Dukʷibəɬ ‘Changer’, thereby highlighting that event-participant for the addressee

and singling out Changer him as the unique individual responsible for the theft.

In all of the sentences in (706) and (707), the adverb diɬ takes scope over the subject of the

sentence, which in all four cases is some kind of anaphoric element whose antecedent is an

event-participant from earlier in the discourse. Even in cases where the subject is an overt NP,

such as (704a) above, this NP is co-referent with some Topical element in the preceding text.

Because of the near-absolute coincidence between subject and Topic/Theme in Lushootseed

discourse (Section 11.3), the scope of diɬ encompasses both the subject of the sentence and the

Theme; however, in those cases where the subject is not Thematic, the scope of the focalizer can

be shown to be determined by Thematicity rather than syntactically, as shown in (708):

(708) a. diɬ čəɬ gʷəkʷədad tiʔiɬ

diɬ čəɬ gʷə=kʷəda–d tiʔiɬ

FOC 1PL.SUB SBJ=taken–ICS DIST

‘that guy’s the one we should get’

(Hess 2006: 18, line 167)

b. gʷəl diɬ čəd ləgʷč’əd

gʷəl diɬ čəd lə=gʷəč’–d–Ø Ø

SCONJ FOC 1SG.SUB PROG=look.for–ICS–3OBJ 3SUB

‘they are the ones we are looking for’

(Bierwert 1996: 221, line 522)

(708a) is taken from a context where two brothers, who are looking for someone to help rescue

their sister who is trapped in a tree, have been discussing the talents of a particular person

516

(ƛ’əƛ’iq’šəd ‘Sapsucker’). The subject of the sentence is the pronominal clitic čəɬ, but the

focalized element is clearly the demonstrative tiʔiɬ, used in this context as a pronominal whose

antecedent is the current discourse Topic, Sapsucker, the person with the requisite abilities to

rescue the sister. Similarly, (708b) occurs at a point in the text where the speaker is recapping an

explanation, given in the two previous lines in response to a question, of who it is in particular

that is the object of his search. The subject of the sentence is the first person, but the focalized

element is the Topical, and Thematic, direct object (in this case, an anaphoric zero) whose

identity has been established in the preceding two lines of text. Thus, it seems that in both cases

the scope of the focalizer is determined by Thematic, rather than syntactic structure.

As well as modifying verbal predicates, diɬ is also very frequently used as a modifier of

nominal predicates (Section 8.3.1) such as those in (709):

(709) a. hay ƛ’ubəxʷ diɬ adčəgʷasəxʷ tsiʔiɬ

hay ƛ’ub=əxʷ diɬ ad–čəgʷas=əxʷ tsiʔiɬ

SCONJ good=now FOC 2SG.PO–wife=now DIST:FEM

‘well then that one is the one who should be your wife’

(Hess 1998: 98, line 201)

b. gʷaʔ diɬ bads tiʔəʔ ʔəshuygʷastxʷ

gʷəl diɬ bad–s tiʔəʔ ʔəs–huy•gʷas–txʷ

SCONJ FOC father–3PO PROX STAT–be.done•pair–ECS

‘as the one to whom she had been married [was] her father’

(Hess 1998: 99, line 224)

c. gʷəl diɬ bədaʔs əlgʷəʔ tiʔiɬ ʔux%aax%aabiʔɬ

gʷəl diɬ bədaʔ–s əlgʷəʔ tiʔiɬ ʔu–x%aa–x%aab•iʔɬ

SCONJ FOC offspring–3PO PL DIST PFV–DSTR–cry•child

‘and it is their baby who cries on and on’

(Hess 2006: 6, line 97)

In the sentences in (709), the first of the two nouns in each sentence functions as the predicate

and the full NP introduced by the determiner is the subject. The focalizer diɬ appears in the

expected position for a lexical adverb, preceding the clausal predicate, and has the effect of

focalizing the Thematic subject. Note that in (709a) diɬ follows (rather than precedes, as in 706b)

517

another lexical adverb, ƛ’ub ‘well’, reflecting the relatively free respective ordering of lexical

adverbs (Section 2.5.1).

Sentences with nominal predicates and focalized Themes tend to occur in contexts where the

narrative highlights a particular person or thing, singling it out for special attention, as does the

sentence in (710d):

(710) a. k’əyiɬəxʷ

k’əyiɬ=əxʷ

pretend=now

‘he pretends’

b. ləsxʷɬil bad

ləs–xʷɬ–il bad

PROG.STAT–lack–INCH father

‘he does not have a father’

c. k’əyiɬəxʷ

k’əyiɬ=əxʷ

pretend=now

‘he pretends’

d. diɬ tubads, tusbiaw, kʷi ʔuʔatəbəd

diɬ tu=bad–s tu=sbiaw kʷi ʔu–ʔatəbəd

FOC PAST=father–3PO PAST=coyote REM PFV–die

‘the one who died, it was his father, Coyote’

(Hess 2006: 31, lines 224 – 227)

The lines in (710) occur at a point in a narrative where a character, Coyote, has faked his own

death and is now impersonating his son (whom he has tricked into becoming lost in the Sky

World). Beginning in (710a), Coyote, thus disguised, puts on a show of mourning for the

supposedly deceased individual, singled out and focalized in (710d) as Coyote, ostensibly the

mourner’s father. By using the adverbial diɬ in this sentence, the narrator converts what would be

a simple statement of fact, ‘the one who died is his father, Coyote’, to a more emphatic statement

identifying the deceased as a particular individual, whose specific identity holds special

significance for the speaker and/or the narrative context.

518

In addition to modifying sentences that have simple nouns as their predicates, diɬ can modify

more complex nominal predicates, such as those in the sentences in (711):

(711) a. gʷəl diɬ dsəsbək’ʷucid tiʔəʔ q’ədᶻax% ʔə tsi sqigʷac

gʷəl diɬ d=s=ʔəs–bək’ʷ•ucid tiʔəʔ q’ədᶻax% ʔə

SCONJ FOC 1SG.PO=NM=STAT–scavenge•mouth PROX intestines PR

tsi sqigʷac

SPEC:FEM deer

‘and the intestines of deer are what I happened upon (that I could eat)’

(ML Mink and Tutyika I, line 180)

b. diɬ shuys

diɬ s=huy=s Ø

FOC NM=made=3PO 3SUB

‘that is the end’

(Hess 1995: 146, line 92)

The predicates of these sentences — in (711a), dsəsbək’ʷucid ‘what I happened upon’ (lit. ‘what

I have scavenged with the mouth’), and in (711b) shuys ‘its ending’ (lit. ‘its being done’) — are

s-nominals. The role of diɬ in both sentences is to focalize the Thematic subject (tiʔəʔ q’ədᶻax& ʔə

tsi sqigʷac ‘Deer’s intestines’ in 711a and the zero pronominal in 711b). The referent of the

pronominal in the latter case is the traditional story the speaker has just told, this being a

common way of terminating a Lushootseed narrative.

Like s-nominals, adjunct nominals are also found as predicates modified by diɬ:

(712) a. diɬ cəxʷəsčəbaʔ tiʔəʔ dsəsc’iqʷib tsiʔəʔ ʔuʔu

diɬ d=dəxʷ=ʔəs–čəbaʔ tiʔəʔ d=s=ʔəs–c’iqʷib

FOC 1SG.PO=ADNM=STAT–laden PROX 1SG.PO=NM=STAT–get.in.on

tsiʔəʔ ʔuʔu

PROX:FEM a.bit

‘what I got in on, this little bit, is what I am loaded down with’

(Hess 1998: 75, line 249)

b. diɬ ɬudəxʷʔəɬəds tiʔiɬ tudəxʷciltubs

diɬ ɬu=dəxʷ=ʔəɬəd=s tiʔiɬ tu=dəxʷ=cil–txʷ–b=s

FOC IRR=ADNM=feed.on=3PO DIST PAST=ADNM=dish–ECS–PASS=3PO

‘that [thing] in which [food] was dished out is what he will eat with’

(Hess 2006: 38, line 409)

519

The predicate of (712a) is the dəxʷ=nominal cəxʷəsčəbaʔ ‘what I am loaded down with’, while

in (712b) the predicate is ɬudəxʷʔəɬəds ‘what he will eat with’. Again, the role of diɬ here is to

focalize and particularize the Theme/subject of the sentence.

As in the above examples, nominal predicates modified by diɬ tend to be bare nominals

without determiners; as with bare nominal predicates; however, it is also possible to find

sentences with diɬ in which the predicate nominal is introduced by a determiner:

(713) a. ʔu, diɬiɬ tiʔiɬ saʔ luƛ’ tiʔiɬ ʔugʷəlald ti tudsč’istxʷ

ʔu diɬ–iɬ tiʔiɬ saʔ luƛ’ tiʔiɬ ʔu–gʷəlald ti tu=d–sč’istxʷ

INTJ INTNS–FOC DIST bad elder DIST PFV–kill SPEC PAST=1SG.PO–husband

‘oh, that very one who killed my husband is that bad old man’

(Hess 2006: 31, line 232)

b. gʷəhaw’ə diɬ tiʔəʔ tusč’istxʷs tiʔəʔ ʔutəɬəɬdubutəxʷ

gʷə=haw’ə diɬ tiʔəʔ tu=sč’istxʷ–s tiʔəʔ

SBJ=PTCL FOC PROX PAST=husband–3PO PROX

ʔu–təɬəɬ–dxʷ–but=əxʷ

PFV–arrive.safely–DC–REFL=now

‘why this very one who safely arrived was her former husband!’

(Hess 2006: 33, line 279)

The conditions determining whether or not the predicate complement takes a determiner are the

same as those governing the presence/absence of a determiner with other predicate nouns

(Section 8.3.1): essentially, the contrast resides in whether the sentence attributes membership in

a type named by the predicate nominal, as in (709), or whether the subject is being identified

with a specific referential entity, as in (713). In either case, the function of diɬ remains the

focalization and particularization of the Thematic subject. When the predicate is a bare nominal,

diɬ is used to indicate that the Theme/subject is the unique member of the type expressed by the

predicate in the current discourse. When the predicate is introduced by the determiner, the role of

diɬ is to emphasize that is it a particular individual — out of all the other potential candidates in

the discourse — that is being identified with entity expressed by that predicate.

David Beck 10-2-7 2:19 PM

Comment: is the RDP intns or exc?

520

Although diɬ can be used in copular constructions with both a predicate nominal and a

subject, diɬ frequently appears with only a predicate nominal, and the elided subject is

interpreted as either a Topical event-participant or as referring to the immediately preceding

events themselves, the elided or zero element acting much the way English this and that do in

spoken and informal written discourse:

(714) a. diɬəxʷ ɬudsəsq’ʷuhuʔ

diɬ=əxʷ ɬu=d=s=ʔəs–q’ʷuhuʔ Ø

FOC=now IRR=1SG.PO=NM=STAT–associate.with 3SUB

‘[she] is the one I will associate with’

(Hess 2006: 31, line 241)

b. diɬ sx%aƛ’s

diɬ s=x%aƛ’=s Ø

FOC NM=want=3PO 3SUB

‘that was what he wanted’

(Hess 1998: 92, line 33)

c. diɬ tushuyutəbsəxʷ ʔə tiʔəʔ dukʷibəɬ

diɬ tu=s=huyu–t–əb=s=əxʷ ʔə tiʔəʔ dukʷibəɬ Ø

FOC PAST=NM=be.done–ICS–PASS=3PO=now PR PROX Changer 3SUB

‘that is what was done to her by Changer’

(Hess 1998: 75, line 256)

d. diɬəxʷ dəxʷʔəsčcils

diɬ=əxʷ dəxʷ=ʔəs–čc–il=s Ø

FOC=now ADNM=STAT–red–INCH=3PO 3SUB

‘that [event] is why he [Kingfisher] is red’

(Hess 2006: 15, line 104)

The subject in (714a) is understood in context to be a character just named in discourse (Sawbill,

the wife of Coyote’s son whom the speaker, Coyote, has lusted after), while the subject of (714b)

is the list of desires (the funeral arrangements being requested by the protagonist, Coyote) just

enumerated by the storyteller. (714c) and (d) make concluding statements about preceding

events, which are themselves the referent of the understood, Focalized subject.

521

11.2.2 Contrastive focalization using gʷəl

The second strategy for overtly marking Focalization is a specialized use of the conjunction

gʷəl which places the focalized element, an NP or PP, at the beginning of the sentence, followed

by gʷəl, as shown in (715):

(715) a. tiʔiɬ sbiaw gʷəl ʔux%ʷəxʷ

tiʔiɬ sbiaw gʷəl ʔux%ʷ=əxʷ

DIST coyote SCONJ go=now

‘Coyote, well, he goes’

(Hess 1995: 148, line 45)

177

b. tiʔəʔ swatixʷtəd gʷəl tasbəsad ʔal kʷi tuhaʔkʷ

tiʔəʔ swatixʷtəd gʷəl tu=ʔas–bəsad ʔal kʷi tu=haʔkʷ

PROX country SCONJ PAST=STAT–dark at REM PAST=long.time

‘the land, well, it was dark in the distant past’

(Hilbert & Hess 1977: 13)

Both of the sentences in (715) have verbal predicates whose subject NP appear clause-initially,

followed by gʷəl, rather than immediately following the subject. gʷəl-focalization is also attested

in sentences with non-verbal predicates as in (716a) — predicated on the negative adverb xʷiʔ —

and (716b), which has a nominal predicate, həlaʔb hikʷ sʔax&ʷuʔ ‘really big clam’:

(716) a. tiʔəʔ swatixʷtəd gʷəl xʷiʔ gʷəsəliʔs

tiʔəʔ swatixʷtəd gʷəl xʷiʔ gʷə=s=həliʔ=s

PROX country SCONJ NEG SBJ=NM=alive=3PO

‘the land, well, it was not alive’

[DS Star Child, line 375]

b. haʔəc gʷəl həlaʔb hikʷ sʔax%ʷuʔ

haʔəc gʷəl həlaʔb hikʷ sʔax%ʷuʔ

horse.clam SCONJ really big clam

‘the horse clam, well, it is a really big clam’

(Hess 1995: 122, ex. 1)

Frequently, sentences with focalized NPs are introduced by an additional instance of gʷəl in its

role as a sentential conjunction (Section 2.7.5):

177

In the published source, this sentence appears with an editorial amendment adding the perfective aspect marker to

the verb, rather than with the verb in the imperfective aspect as it is spoken on tape. As either version would be

correct, the original sentence from the recording is presented here.

522

(717) a. gʷəl tiʔəʔ qaw’qs gʷəl ƛ’al’ ʔəbsbibədbədaʔ ʔə tiʔəʔ bəqa

gʷəl tiʔəʔ qaw’qs gʷəl ƛ’al’ ʔəs–bəs–bi–bəd–bədaʔ

SCONJ PROX raven SCONJ also STAT–PROP–ATTN–DSTR–offspring

ʔə tiʔəʔ bə=qa

PR PROX ADD=many

‘and Raven, well, he also had children aplenty’

(Hess 1998: 78 line 8)

b. gʷəl tsiʔəʔ ʔiɬt’isuʔ gʷəl ƛ’uləqdxʷ tsiʔiɬ luƛ’

gʷəl tsiʔəʔ ʔiɬ–t’isuʔ gʷəl ƛ’u=ləq–dxʷ tsiʔiɬ luƛ’

SCONJ PROX:FEM PRTV–younger SCONJ HAB=listen–DC DIST:FEM old

‘and the younger one, well, she overheard the old woman’

[DS Star Child, line 123]

In terms of textual frequency, the pattern shown in (717) is actually more common for sentences

with focalized NPs than the simpler construction in (715). This is probably due to the

coincidence of the role gʷəl-focalization plays in topic-setting (see 11.3.1 below) with the use of

sentential conjunctions like gʷəl in the setting of episode and other boundaries in discourse

(Cook 1999).

When the focalized element is an anaphor, it is expressed as a lexical pronoun (Section

2.7.1), as in (718):

(718) a. gʷəl cədiɬ, gʷəl x%ʷul’ ʔup’ayəq

gʷəl cədiɬ gʷəl x%ʷul’ ʔu–p’ayəq

SCONJ he SCONJ only PFV–carve.canoe

‘him, well, he just carved canoes’

(Hess 2006: 42, line 8)

b. gʷəl dəgʷi gʷəl ʔahəxʷ sixʷ gʷadsudᶻaldᶻalalus

gʷəl dəgʷi gʷəl ʔa=əxʷ sixʷ gʷə=ad=s=ʔu–dᶻal–dᶻal•alus

SCONJ you SCONJ be.there PTCL SBJ=2SG.PO=NM=PFV–DSTR–turn•eye

‘and you, well, you’d [still] be there going from end to end [of the canoe]’

[ML Mink and Tutyika I, line 114]

c. dəgʷiˑ gʷəl dəxʷəliʔ čəɬ, dəxʷp’al’p’alil čəɬ

dəgʷi gʷəl dəxʷ=həliʔ čəɬ dəxʷ=p’al’–p’alil čəɬ

you SCONJ ADNM=alive 1PL.PO ADNM=DSTR–revive 1PL.PO

‘you, well you’re the reason we are alive, why we have revived’

(Hess 2006: 78, line 883)

523

As shown by the examples in (718b) and (c), the expression of the subject of the sentence as a

fronted NP suppresses the expected expression of the subject as a matrix subject clitic —

compare, for instance, (718b) and (719), which are syntactically parallel except in that the

second-person subject is focalized in the former case but not in the latter:

(719) x%ʷul’əxʷ čəxʷ ɬuʔa kʷi ɬadsucucut ʔal kʷi ɬuč’itəxʷ ɬuʔaciɬtalbixʷ

x%ʷul’=əxʷ čəxʷ ɬu=ʔa kʷi ɬu=ad=s=ʔu–cut–cut

only=now 2SG.SUB IRR=be.there REM IRR=2SG.PO=NM=PFV–DSTR–say

ʔal kʷi ɬu=č’it=əxʷ ɬu=ʔaciɬtalbixʷ

at REM IRR=near=now IRR=people

‘you will simply be there talking away when the future people arrive’

(Hess 2006: 7, line 117)

Here, the second-person subject — fronted because of the appearance of the adverb x&ʷul’ ‘only’

— continues to be expressed by a subject clitic, as opposed to its expression by the lexical

pronoun dəgʷi in (718b), indicating that focalization of the subject with gʷəl requires its

expression as a pronominal.

In addition to focalizing NPs, gʷəl can be used to focalize adverbial prepositional phrases,

such as those in (720):

(720) a. ʔal suʔəɬəds ʔə tiʔiɬ sʔuladxʷs gʷəl xʷiˑʔ kʷi gʷəsbək’ʷdxʷs

ʔal s=ʔu–ʔəɬəd=s ʔə tiʔiɬ sʔuladxʷ–s gʷəl xʷiʔ kʷi

at NM=PFV–feed.on=3PO PR DIST salmon–3PO SCONJ NEG REM

gʷə=s=bək’ʷ–dxʷ=s

SBJ=NM=all–DC=3PO

‘as he ate the salmon, well, he couldn’t manage to eat anymore’

(Hess 1995: 152, line 23)

b. ʔaləxʷ sɬixʷəɬdacuts kʷi sʔaʔs bibščəb gʷəl ʔux%ʷc tiʔəʔ sləx%il

ʔal=əxʷ s=ɬixʷ•əɬ•dat–t–sut=s kʷi s=ʔa=s bi–bščəb

PR=now NM=three•CLS•day–ICS–REFL=3PO REM NM=be.there=3PO ATTN–mink

gʷəl ʔux%ʷ–c tiʔəʔ sləx%il

SCONJ go–ALTV PROX day

‘on the third day that Little Mink is there, well, he goes after the daylight’

(Hilbert & Hess 1977: 23)

524

c. ʔal kʷi tuhaʔkʷ gʷəl xʷiʔ gʷətusləx%il

ʔal kʷi tu=haʔkʷ gʷəl xʷiʔ gʷə=tu=sləx%il

PR REM PAST=long.time SCONJ NEG SBJ=PAST=day

‘in the distant past, well, there was no daylight’

(Hilbert & Hess 1977: 13)

As with NPs, the focalization of PPs places the adverbial at the beginning of the sentence

followed by gʷəl and then the sentence predicate. Focalized PPs are frequently attested in a

scene-setting function (Section 11.3.1).

The use of gʷəl discussed here overlaps in its use a great deal with the other focalizer, diɬ,

although generally-speaking the two differ in that gʷəl-focalization is more directly contrastive

and tends to fill a topic- or scene-setting function, whereas diɬ is more a particularizer that

singles out a unique individual or event, and is not typically used to introduce new topics. There

are, however, a few examples of copular constructions that make use of both focalizers, such as

those in (721):

(721) a. gʷəl tsiʔəʔ həbuʔ gʷəl diɬ čəgʷas

gʷəl tsiʔəʔ həbuʔ gʷəl diɬ čəgʷas–s

SCONJ PROX:FEM pigeon SCONJ FOC wife–3PO

‘and Pigeon, well, she was his true wife’

(Hess 2006: 22, line 14)

b. tsiʔiɬ cədiɬ gʷəl diɬ bədaʔs

tsiʔiɬ cədiɬ gʷəl diɬ bədaʔ–s

DIST:FEM he SCONJ FOC offspring–3PO

‘her, she was the one who was her daughter’

[ML Basket Ogress, line 27]

Although these constructions are rare, it seems that their primary function is to combine the

contrastive and particularizing functions of the two focalizing constructions. Thus, (721a) comes

at a point in narrative where the storyteller is discussing the merits and shortcomings of the two

wives of Coyote’s son, Sawbill and Pigeon. Sawbill is beautiful and is lusted after by her father-

in-law, Coyote, but — as (721b) makes clear — in contrast to Sawbill, Pigeon and Pigeon alone

is Coyote’s son’s true wife. The contrast with Sawbill is signaled by gʷəl-focalization, while the

525

remainder of the sentence — which literally translated would simply mean ‘she was his wife’ —

derives the particularizing sense of ‘true wife’ conveyed in the gloss through the use of diɬ.

11.3 Topic-alignment and reference-tracking

In addition to the strong preference for Thematic subjects, one of the principal organizing

features of Lushootseed discourse is a constraint that the syntactic subject of a clause, which is

generally Thematic, also be discourse Topic.

178

This constraint plays a crucial role both in the

interpretation of clauses within a local context and in the structuring of conversation and

narratives into episodes organized around a shared Thematic subject — that is, a discourse

Topic. At the level of the individual clause, the alignment of subject and Topic ensures the

unambiguous interpretation of sentences and stretches of discourse, even when these are third-

persons expressed as zero subject pronominals; at the level of the narrative or text, the Topic-

driven organization of Lushootseed discourse results in the arrangement of sets of utterances or

lines into narrative or conversational episodes which share a common Thematic subject, as in

(722):

(722) a. ɬčiləxʷ tsiʔəʔ sɬadəyʔ

ɬčil=əxʷ tsiʔəʔ sɬadəyʔ

arrive=now PROX:FEM woman

‘the woman (Basket Ogress) arrives’

b. ʔa haw’əʔ ʔal tiʔiɬ

ʔa haw’əʔ Ø

ʔal tiʔiɬ

be.there PTCL 3SUB at DIST

she

is right there’

c. xʷiʔ gʷəsəsaydubs ʔə tiʔəʔ diʔəʔ

xʷiʔ gʷə=s=ʔəs–hay–dxʷ–b=s

ʔə tiʔəʔ diʔəʔ

NEG SBJ=NM=STAT–KNOWN–DC–PASS=3PO PR PROX he

‘she

wasn’t known (i.e., seen) by the children’

178

This pattern is the predominant pattern in the Salishan family in general, as discussed in detail for a variety of

languages, including Lushootseed, in Kinkade (1990).

526

d. ɬčiləxʷ

ɬčil=əxʷ Ø

arrive=now 3SUB

‘she

arrives’

e. gʷəl kʷədadəxʷ tiʔəʔ stawixʷaʔɬ

gʷəl kʷəda–d–Ø=əxʷ Ø

tiʔəʔ stawixʷaʔɬ

SCONJ taken–ICS–OBJ=now 3SUB PROX children

‘then she

took the children

’

f. kʷədabidəxʷ

kʷəda–bi–d–Ø=əxʷ Ø

taken–MAP–ICS–3OBJ=now 3SUB

‘she

captured them

’

g. kʷədad tiʔəʔ

kʷəda–d–Ø Ø

tiʔəʔ

taken–ICS–3OBJ 3SUB PROX

‘she

took them

’

h. gʷəl lədəgʷaš

gʷəl lə=dəgʷa–š–Ø Ø

SCONJ PROG=inside–ICS–3OBJ 3SUB

‘she

is putting them

inside (her basket)’

i. lədəgʷaš dxʷʔal tiʔəʔ xʷʔax%ʷaʔads

lə=dəgʷa–š–Ø Ø

dxʷ–ʔal tiʔəʔ xʷʔax%ʷaʔad–s

PROG=inside–ICS–3OBJ 3SUB CNTRFG–at PROX clam.basket–3PO

‘she

is putting them

inside her clam basket’

j. ʔahiləxʷ tiʔəʔ kikəwič

ʔa–il=əxʷ tiʔəʔ ki–kəwič

be.there–INCH=now PROX ATTN–hunchback

‘Little Hunchback ends up in there’

[MS Basket Ogress, lines 5 – 14]

The identity of the Topical event-participant, Basket Ogress, is expressed overtly only once in

the first nine lines of text here, in (722a). Nevertheless, the interpretation of these sentences is

unambiguous: once Basket Ogress is established as the discourse Topic, the alignment of Topic

and subject ensures that this event-participant be interpreted as the subject of all of the sentences

in the ensuing discourse. Thus, the subject of (722a) is the antecedent of the zero subject

pronominal subject in (722b), the third-person possessive subject in (722c), and the zero subject

527

of the intransitive verb in (722d). It is also the antecedent of the zero subject pronominal of the

transitive sentence in (722e), which introduces a new event-participant into the episode, tiʔəʔ

stawixʷaʔɬ ‘the children’ as well. Even in the presence of the new event-participant, Basket

Ogress remains the Topic of this episode and continues to be the antecedent of the zero subject

anaphora in subsequent lines such as (722f). Even though both subject and object are zeroes in

this and the next lines of the episode, their subjects refer back unambiguously to Basket Ogress,

while their direct object is ‘the children’. This pattern is maintained up until line (722j), where a

sentence with another overt subject NP — tiʔəʔ kikəwič ‘Little Hunchback’ — appears,

supplanting Basket Ogress and setting the Topic for the subsequent discourse episode.

One important consequence of the need to align subject and Topic over specific discourse

episodes is the frequent use of the passive voice to maintain Topical event-participants in subject

position when they do not correspond to the subject of the active form of a particular verbal

predicate. In following passage from the opening of “bibščəb ʔi tiʔiɬ suʔsuq’aʔs, tətyika” ‘Little

Mink and his younger cousin, Tutyika’ as told by Mr. Edward Sam, for example, the speaker

shifts into the passive voice mid-way through the episode as a new, non-topical event-participant

is introduced and becomes the initiator of the subsequent action:

(723) a. hay, ʔuɬiʔɬdahəb tiʔiɬ bibščəb ʔi tiʔiɬ suʔsuq’ʷaʔs, tətyika

hay ʔu–ɬiʔɬdahəb tiʔiɬ bi–bščəb ʔi tiʔiɬ suʔ–suq’ʷaʔ–s tətyika

SCONJ PFV–troll DIST ATTN–mink and DIST ATTN–younger.cousin Tutyika

‘well then, Little Mink and his younger cousin, Tutyika

, went trolling’

b. ʔuɬiʔdaab əlgʷəʔ

ʔu–ɬiʔdaab Ø

əlgʷəʔ

PFV–troll 3SUB PL

‘they

went trolling’

c. huy, šudxʷəxʷ tiʔiɬ čxʷəluʔ

huy šuɬ–dxʷ–Ø=əxʷ Ø

tiʔiɬ čxʷəluʔ

SCONJ see–DC–3OBJ=now 3SUB DIST whale

‘well, they

caught sight of Whale

’

528

d. huy, bapad$xʷ əlgʷəʔ

huy bapa–d–Ø=əxʷ Ø

əlgʷəʔ

SCONJ annoyed–ICS–3OBJ=now 3SUB PL

‘well, they

annoyed him

’

e. bapad$xʷ əlgʷəʔ tiʔiɬ čxʷəluʔ

bapa–d–Ø=$xʷ Ø

əlgʷəʔ tiʔiɬ čxʷəluʔ

annoyed–ICS–3OBJ=now 3SUB PL DIST whale

‘they

annoyed that Whale

’

f. huy, xʷak’ʷisəbəxʷ ʔə tiʔiɬ čxʷəluʔ

huy xʷak’ʷil–s–b=əxʷ Ø

ʔə tiʔiɬ čxʷəluʔ

SCONJ get.sick.of–ALTV–PASS=now 3SUB PR DIST whale

‘well, they

were gotten sick of by that Whale

’

g. huy, bəq’təbaxʷ ʔə tiʔiɬ čxʷəluʔ

huy bəq’–t–b=axʷ Ø

ʔə tiʔiɬ čxʷəluʔ

SCONJ be.in.mouth–ICS–PASS=now 3SUB PR DIST whale

‘well, they

were swallowed by that Whale

’

(Hess 1995: 140, lines 6 – 12)

The narrator begins in (723a) by setting a discourse Topic — tiʔiɬ bibščəb ʔi tiʔiɬ suʔsuq’ʷaʔs,

tətyika ‘Little Mink and his younger cousin, Tutyika’ — which becomes the antecedent for the

zero subject anaphora of the sentences throughout the remainder of the episode. In sentence

(723c) a new participant, Whale, is introduced as the direct object of the verb šudxʷ ‘catch sight

of somebody’. The next line (723d) has both a zero subject and direct object, but is unambiguous

as Little Mink and his cousin remain the discourse Topic and so are interpreted as the

subject/AGENT of the verb bapad ‘annoy somebody’. Line (723d) is repeated in line (723e) with

an overt object NP (tiʔiɬ čxʷəluʔ ‘Whale’). Up to this point, the actions in the narrative have

been initiated by the topical participants, Little Mink and his cousin, but in (723f) the protagonist

of the event is Whale, which would normally be the subject of the verb xʷak’ʷis ‘get sick of

somebody’. However, in order to avoid violating Topic–subject alignment, the speaker uses the

passive voice — xʷak’ʷisəb ‘be gotten sick of’, thereby allowing the topical event-participants,

Little Mink and his cousin, to be maintained in subject position. The same thing happens in

(723g), where the AGENT of the action, bəq’əd ‘swallow something’, is the non-topical event-

529

participant, Whale, again requiring the passive to promote the topical Little Mink and his cousin

to subject and to maintain the necessary alignment.

The particular type of textual organization illustrated in (722) and (723) is, of course, but one

(and the simplest) many variations on the same basic pattern — the instantiation of a single

central (and usually animate) event-participant as discourse Topic over a short stretch of text. In

conversation and discourse, speakers frequently describe situations involving multiple

participants over episodes that are dozens of lines long, and manipulate their discourse saliency

and Topicality in complex ways that are poorly understood even in familiar languages like

English, and whose elucidation in Lushootseed would be well beyond the scope of this grammar.

The organization of discourse at this level in fact interacts with and depends on a wide variety of

factors including rhetorical and stylistic considerations, manipulation of the shared knowledge

and cultural expectations of speaker and addressee, and individual story-teller’s art,

179

much of

which is regulated by processes well outside of the morphosyntactic machinery of a language.

There are, however, certain features of the morphosyntax that do serve, as either their primary or

secondary function, to encode specific features of discourse. Even if their functions belong more

to the realm of text-planning than to that of morphosyntax, their form is by nature

morphosyntactic, and so merit more detailed discussions in the sections below.

11.3.1 Topic- and scene-setting constructions

Because of the importance of Topics for the organization of discourse into episodes and for

the interpretation of sentences in context, in texts it is often necessary to highlight event-

participants that will become the discourse Topic for the subsequent episode — that is, to

topicalize them. While Lushootseed lacks any one specific morphosyntactic construction

179

On these topics, see Bierwert (1993, 1996), Langen (1997, 1998a, 1998b, 1999), and Moses and Langen (2001).

530

dedicated to this kind of topicalization, it does employ a variety of different constructions in this

function:

Non-verbal predicates

Non-verbal predicates are frequently found at the beginning of discourse episodes, their

Rhematic portion becoming the Topic for subsequent lines of text. This is especially true of

nominal predicates, which are frequently used to introduce new event-participants or previously-

introduced event participants in novel relationships to events or other participants in the

discourse. Consider (724b), which is from a discourse episode that introduces a new participant

in discourse, tiʔiɬ buʔqʷ ‘the Duck People’:

(724) a. gʷəcutadəxʷ kʷədiʔ tul’səqʷ

gʷəcutad=əxʷ kʷədiʔ tul’–səqʷ

be.commotion=now DIST.DMA CONTRFG–high

‘there was a commotion up there’

b. balgʷas buʔqʷ kʷi bək’ʷ sčads ʔə ti buʔqʷ gʷəɬ x%ʷəlč, tiʔiɬ bəgʷəɬ t’aq’t, gʷəɬ spaɬx%ad

buʔqʷ

balgʷas buʔqʷ Ø kʷi bək’ʷ s=čad=s ʔə ti buʔqʷ

all.kinds waterfowl 3SUB REM all NM=where=3PO PR SPEC waterfowl

gʷəɬ x%ʷəlč tiʔiɬ bə=gʷəɬ t’aq’t gʷəɬ spaɬx%ad buʔqʷ

belong sea DIST ADD=belong inland belong tidal.flats waterfowl

‘it was all kinds of Duck People from everywhere belonging to the sea, and belonging

to the mountains, belonging to the tidal flats’

b. šidᶻəxʷ tiʔiɬ buʔqʷ

šidᶻ=əxʷ tiʔiɬ buʔqʷ

attack=now DIST waterfowl

‘the waterfowl attack by stealth’

c. gʷəl šidᶻəxʷ

gʷəl šidᶻ=əxʷ Ø

SCONJ attack=now 3SUB

‘they attacked by stealth’

(Hess 2006: 61, lines 456 – 459)

Previous to these lines, the narrator has related that the protagonists (two brothers who have been

captured by the Dwarves) are in the long house of the Dwarves when they hear a commotion on

531

the roof (724a). The next line, (724b), reveals the identity of those making the noise, using a

nominal predicate construction in which the people being identified (the Duck People) are the

predicate and are introduced into discourse as new event-participants. This predicate then

becomes the Topic and syntactic subject of the following lines of text, beginning in line (724c),

which repeats the new Topic as an overt NP subject. This strategy of re-enforcing the Topicality

of an event-participant introduced as a nominal predicate is also seen in (725):

(725) a. tiʔiɬ bibščəb ʔi tiʔiɬ suʔsuq’ʷaʔs, tətyika, tiʔiɬ ɬudsyəhubtubicid

tiʔiɬ bi–bščəb ʔi tiʔiɬ suʔ–suq’ʷaʔ–s tətyika,

DIST ATTN–mink and DIST ATTN–younger.cousin–3PO Tutyika

tiʔiɬ ɬu=d=s=yəhub–txʷ–bicid

DIST IRR=1SG.PO=NM=tell–ECS–2SG.OBJ

‘what I will tell you about [is] Little Mink and his younger cousin, Tutyika’

b. hay, ʔuɬiʔɬdahəb tiʔiɬ bibščəb ʔi tiʔiɬ suʔsuq’ʷaʔs, tətyika

hay ʔu–ɬiʔɬdahəb tiʔiɬ bi–bščəb ʔi tiʔiɬ suʔ–suq’ʷaʔ–s tətyika

SCONJ PFV–troll DIST ATTN–mink and DIST ATTN–younger.cousin Tutyika

‘well then, Little Mink and his younger cousin, Tutyika, went trolling’

(Hess 1995: 140, lines 5 – 6)

Line (725a), which immediately precedes the full discourse episode given in (723) above,

identifies quite explicitly for the addressee the main characters of the narrative, making these —

tiʔiɬ bibščəb ʔi tiʔiɬ suʔsuq’ʷaʔs, tətyika ‘Little Mink and his younger cousin, Tutyika’— the

predicate of the clause, the subject being the headless relative construction tiʔiɬ

ɬudsyəhubtubicid ‘what I am going to tell you’. The new, Rhematic participants in (725a) are

repeated, verbatim, as the subject of (725b) and then become the antecedents of a long string of

zero subjects over the next several lines of text (see 723 above). Although nominal predicate

constructions are found on their own serving a topic-setting function, the pattern in (724) and

(725), whereby the predicate of the non-verbal construction is repeated — either in paraphrase or

verbatim — as the subject of the following line is quite common, and frequently serves to

532

distinguish a nominal predicate in its topic-setting function from a nominal predicate performing

some other role in discourse.

Other types of non-verbal predicates, in particular those expressing locations or times, serve

a similar function, though in these cases they might be better considered scene-setting rather than

topic-setting constructions. Consider, for example, line (726a):

(726) a. ɬixʷəɬdat tiʔiɬ sdəgʷabacilsəxʷ əlgʷəʔ ʔə tiʔiɬ čxʷəluʔ

ɬixʷəɬdat tiʔiɬ s=dəgʷ•abac–il=s=əxʷ əlgʷəʔ ʔə tiʔiɬ čxʷəluʔ

three.days DIST NM=be.inside•body–INCH=3PO=now PL PR DIST whale

‘they are inside that whale for three days’

(lit. ‘their being inside that whale [is for] three days’)

b. huy, ʔibibišəxʷ tiʔiɬ bibščəb

huy ʔib–ʔibiš=əxʷ tiʔiɬ bi–bščəb

SCONJ DSTR–travel DIST ATTN–mink

‘then Little Mink paces back and forth’

(Hess 1995: 140, lines 13–14)

This line of text immediately follows the discourse episode in (723), at the end of which the

protagonists, Little Mink and his cousin, are swallowed by Whale. By explicitly stating how long

it is that Mink and his cousin are inside Whale in (726a), the scene is set for the ensuing action,

beginning in (726b), which itself sets the discourse Topic for the rest of the episode (tiʔiɬ

bibščəb ‘Little Mink’ — see 729 below).

gʷəl-focalization

As described in Section 11.2.2 above, gʷəl-focalization involves the fronting of a Thematic

sentence constituent, which is then followed by gʷəl and becomes the Focalized portion of the

utterance. In terms of their uses in discourse, constructions of this type with gʷəl typically serve

one of two functions, depending on the syntactic type of the fronted constituent — that is,

whether it is an NP or a PP. In the former case, the fronted NP seems to be used in a contrastive

way, in contexts where the speaker is emphasizing that the statement being made applies to one

participant as opposed to some other participant present in the current discourse. An example of

533

this can be found at the beginning of the story “tiʔiɬ sčətxʷəd ʔi tsiʔiɬ ƛ’aƛ’ac’apəd” ‘Black Bear

and Ant’ as told by Edward Sam. As indicated in its title, this story involves two main characters,

introduced in the first line of the story:

(727) a. hay, ʔa tiʔəʔ syəyəhub ʔə tiʔiɬ sčətxʷəd ʔi tsiʔiɬ ƛ’aƛ’ac’apəd

hay ʔa tiʔəʔ syəyəhub ʔə tiʔiɬ sčətxʷəd ʔi tsiʔiɬ ƛ’aƛ’ac’apəd

SCONJ be.there PROX legend PR DIST bear and DIST:FEM ant

‘so, there is a traditional story about Black Bear and Ant’

b. tiʔəʔ sčətxʷəd gʷəl x%ʷuˑl’ ƛ’uʔibʔibəš

tiʔəʔ sčətxʷəd gʷəl x%ʷul’ ƛ’u=ʔib–ʔibəš

PROX bear SCONJ only HAB=DSTR–travel

‘this Black Bear, well, he would just amble around’

c. x%ʷuˑl’ ƛ’uʔibʔibəš

x%ʷul’ ƛ’u=ʔib–ʔibəš

only HAB=DSTR–travel

‘he would just wander around’

d. gʷəl tsiʔiɬ ƛ’aƛ’ac’apəd gʷəl dᶻəgʷaʔ dxʷʔulus

gʷəl tsiʔiɬ ƛ’aƛ’ac’apəd gʷəl dᶻəgʷaʔ dxʷʔulus

SCONJ DIST:FEM ant SCONJ expert steady.worker

‘and Ant, well she was a great one for working’

(Hess 1995: 143, lines 1 – 4)

After establishing the identities of the two main characters in line (727a), the narrator goes on to

introduce those character’s defining characteristics — Black Bear is shiftless and lazy, while Ant

is industrious and hard-working. Thus, line (727b) uses a focalizing construction with gʷəl to

introduce Black Bear as the Theme of this and the subsequent line of text in contrast to Ant; in

the next line following these, (727d), the narrator shifts the focus to Ant, whose characteristics

are set up in contrast to those of Black Bear. Because of its contrastive function, gʷəl-

focalization of NPs is used very frequently as a topic-setting construction in this way — in

contexts where the new topic being introduced is one of two or more characters that have been

previously introduced in a stretch of discourse, and the narrator is taking pains to clarify which

of these is the topic of the subsequent episode.

534

Sentences with fronted PPs, on the other hand, often fill the discourse function of scene-

setting in that the focalized adverbial specifies the spatial or (more often) temporal location for

the action of the subsequent discourse, as in the following excerpt from “Stealing Daylight” as

told by Harry Moses. These lines come at a point in the story where Mink is pretending to be an

old man and, covered in ashes, takes up a post by the door of the longhouse where daylight is

kept, waiting for his opportunity to steal it:

(728) a. ʔəšuucəb ʔə ti luƛ’ ʔəsbəč ləq’aɬucid ʔəspədpədič ʔə ti q’ʷaɬčup

ʔəs–šuu–c–b ʔə ti luƛ’ ʔəs–bəč ləq’aɬ•ucid

STAT–see–ALTV–PASS PR SPEC old STAT–lie on.the.way•mouth

ʔəs–pəd–pəd•ič ʔə ti q’ʷaɬ•čup

STAT–DSTR–buried•covering PR SPEC ash•fire

‘the old fellow who is lying by the door covered in fine ash powder watches it’

b. ʔal bəƛ’uɬax% gʷəl bəʔux%ʷcəb ʔə bibščəb

ʔal bə=ƛ’u=ɬax% gʷəl bə=ʔux%ʷ–c–b ʔə bi–bščəb

PR ADD=HAB=dark SCONJ ADD=go–ALTV–PASS PR ATTN–mink

‘when it would get dark again, well, Little Mink would go after it again’

c. buusəɬdat kʷi suʔux%ʷcs tiʔəʔ sɬəx%il gʷəl tuɬaʔačiʔbid

buus•əɬ•dat kʷi s=ʔu–ʔux%ʷ–c=s tiʔəʔ sɬəx%il gʷəl

four•CLS•day REM NM=PFV–go–ALTV=3PO PROX day SCONJ

tu=ɬaʔ•ačiʔ–bi–d

PAST=arrive•hand–MAP–ICS

‘on the fourth day he goes after the daylight, and his hand reached it’

d. diɬ tuspəkʷibids tiʔəʔ sɬəx%il gʷəl tusaxʷəbtxʷ

diɬ tu=s=pəkʷi–bi–d=s tiʔəʔ sɬəx%il gʷəl tu=saxʷəb–txʷ

FOC PAST=NM=snatch–MAP–ICS=3PO PROX day SCONJ PAST=run–ECS

‘thus it was that he snatched daylight from them and ran off with it’

(Hilbert & Hess 1977: 24)

(728a) is the final line of the preceding discourse episode, which describes Mink’s first failed

attempt to get his hands on the prize. The next line begins with a focalized PP setting the time for

the next series of events, the successful theft of daylight, which then becomes the subject-matter

of the following discourse episode. In such cases, the focalize PP is not in fact the Topic of

ensuing discourse, but its setting, the backdrop against which the subsequent action takes place.

535

Prosodic phrasing and F0 reset

Although not strictly-speaking morphosyntactic strategies for topic-setting, prosodic and

intonational processes are also employed for this purpose. One common way of using prosody to

indicate a new discourse Topic is shown in (729a) in which the topicalized element is set off

from the rest of the utterance by an intonational boundary (§), usually one that interrupts the

normal patterns of prosodic phrasing (indicated here by parentheses):

(729) a. ( huy ), ( ʔibibišəxʷ tiʔiɬ ) § ( bibščəb )

huy ʔib–ʔibiš=əxʷ tiʔiɬ bi–bščəb

SCONJ DSTR–travel DIST ATTN–mink

‘then Little Mink paces back and forth’

b. ( ʔibibišəxʷ )

ʔib–ʔibiš=əxʷ Ø

DSTR–travel=now 3SUB

‘he paces’

c. ( huy ), ( k’awdxʷəxʷ ) ( tiʔiɬ sc’aliʔə ) ( tiʔiɬ čxʷəluʔ )

huy k’aw–dxʷ=əxʷ Ø tiʔiɬ sc’aliʔ ʔə tiʔiɬ čxʷəluʔ

SCONJ bump–DC=now 3SUB DIST heart PR DIST whale

‘then he bumps into Whale’s heart’

(Hess 1995: 140, line 14 – 16)

These lines (which follow the lines in 723 and 726 above in the same narrative) recount what

Mink does once he and his cousin have been swallowed by Whale (specifically, they find

Whale’s heart and start a fire next to it, killing him). Line (729a) marks the beginning of the

discourse episode and consists of a full sentence containing an overt NP subject, tiʔiɬ bibščəb

‘Little Mink’, which is the antecedent of the zero subject NPs in the following lines. Its status as

a new discourse Topic is signaled here by the insertion of an intonational boundary between tiʔiɬ

and bibščəb, grouping the determiner into the preceding prosodic phrase (indicated by

parentheses) rather than grouping it with bibščəb, which would be the expected phrasing (cf. the

prosodic groupings in 729c).

180

The use of marked phonological phrasing thus sets apart overt

180

The patterns of Lushootseed prosody and the principles governing prosodic phrasing are discussed in Beck

(1999).

536

subjects used for topic-setting purposes from those introduced into discourse purely for reasons

of disambiguation or reference-tracking.

Not surprisingly, marked intonational boundaries are routinely combined with the other more

overtly marked topic-setting strategies discussed above, as shown in (730):

(730) a. ( tiʔiɬ bibščəb ʔi ) ( tiʔiɬ suʔsuq’ʷaʔs), (tətyika), § ( tiʔiɬ ɬudsyəhububicid )

tiʔiɬ bi–bščəb ʔi tiʔiɬ suʔ–suq’ʷaʔ–s tətyika,

DIST ATTN–mink and DIST ATTN–younger.cousin–3PO Tutyika

tiʔiɬ ɬu=d=s=yəhub–txʷ–bicid

DIST IRR=1SG.PO=NM=tell–ECS–2SG.OBJ

‘what I will tell you about [is] Little Mink and his younger cousin, Tutyika’

(Hess 1995: 140, line5)

b. ( tiʔiɬ sbiaw ) § ( gʷəl ʔux%ʷəxʷ )

tiʔiɬ sbiaw gʷəl ʔux%ʷ=əxʷ

DIST coyote SCONJ go=now

‘Coyote, well, now he goes’

(Hess 1995: 148, line 45)

c. ( hay ), ( c’əldub ) ( tiʔiɬ ) § ( sčətxəd ʔə ) ( tiʔəʔ c’ic’ix% )

hay c’əl–dxʷ–b tiʔiɬ sčətxəd ʔə tiʔəʔ c’ic’ix%

SCONJ defeated–DC–PASS DIST bear PR PROX fish.hawk

‘and so then Black Bear was defeated by Fish Hawk’

(Hess 1995: 154, line 93)

(730a) has a non-verbal predicate which is offset from the subject NP by an intonational

boundary; this predicate, the Rhematic portion of the sentence, then becomes the discourse Topic

for the ensuing episode (see 725 above). In a similar vein, (730b) illustrates a case of gʷəl-

focalization in which the fronted, topicalized NP is set off from the sentence predicate, while

(730c) shows a passive sentence with the marked Subject >> AgCo order. In this case, the

topicalized portion of the sentence is the Theme of the utterance, tiʔiɬ sčətxəd ʔə tiʔəʔ c’ic’ix&

‘Black Bear by Fish Hawk’, which encompasses both the subject and the agentive complement

of the passive verb c’əldub ‘be defeated’. These elements are separated by an intonational

boundary which also occurs in a marked position, separating sčətxəd ‘Black Bear’ from its

determiner and violating the normal pattern of phonological phrasing. Although the combination

537

of prosodic and morphosyntactic strategies for topic-setting is common, it is not absolute: all of

these constructions can be found in text without marked intonation, and marked intonational

contours can sometimes be found with them (in particular, with non-verbal predicate

constructions) when they are not being used for topic-setting.

A shift in discourse Topic is also frequently marked by ultra-high F0 reset (Beck & Bennett

2007). Declination in fundamental frequency is used in Lushootseed narrative as a means of

grouping sentences into discourse episodes, while the reset of the declination marking the

boundaries of narrative episodes and signaling such things as change of scene and change of

action. The example in (731) illustrates F0 reset (↑) accompanying topic-shift and marking

episodic boundaries:

(731) a. ↑ huy ʔibəšəxʷ tiʔəʔ sgʷəlub

huy ʔibəš=əxʷ tiʔəʔ sgʷəlub

SCONJ travel=now PROX pheasant

‘then Pheasant traveled’

b. ʔi ʔibəšəxʷ dxʷčad

ʔi ʔibəš=əxʷ Ø dxʷ–čad

INTJ travel=now 3SUB CNTRFG–where

‘indeed, he traveled everywhere’

c. paƛ’aƛ’ ʔuʔibibəš

paƛ’aƛ’ ʔu–ʔib–ibəš Ø

worthless PFV–DSTR–travel 3SUB

‘he wandered around’

d. x%əɬ ti ƛ’astagʷəxʷ əlgʷəʔ

x%əɬ ti ƛ’u=ʔas–tagʷəxʷ Ø əlgʷəʔ

seemingly HAB=STAT–hungry 3SUB PL

‘it seems they (his family) were always hungry’

e. ↑ diiɬəxʷ kʷi sʔadᶻqdxʷs tiʔəʔ

diʔiɬ=əxʷ kʷi s=ʔadᶻq–dxʷ=s tiʔəʔ

suddenly=now REM NM=meet–DC=3PO PROX

‘suddenly he met them’

538

f. ʔəsgʷaadil tiʔacəc ʔaciɬtalbixʷ

ʔəs–gʷaad–il tiʔacəc ʔaciɬtalbixʷ

STAT–down–INCH UNQ people

‘these people were sitting there’

g. səsaʔliʔ tiʔəʔ ʔaciɬtalbixʷ huy § dxʷləgʷləgʷəb

sə–saʔliʔ tiʔəʔ ʔaciɬtalbixʷ huy dxʷləgʷ–ləgʷəb

HMN–two PROX people SCONJ DSTR–youth

‘there were two people and (they were) youths’

h. gʷəl ʔəbsqʷəbqʷəbayʔ əlgʷəʔ ʔə tiʔəʔ bəsaliʔ

gʷəl ʔəs–bəs–sqʷəb–qʷəbayʔ Ø əlgʷəʔ ʔə tiʔəʔ bə=saliʔ

SCONJ STAT–PROP–DSTR–dog 3SUB PL PR PROX ADD=two

‘and they have two dogs, too’

i. ↑ tiləbəxʷ ʔudxʷsx%ʷuƛ’utəbəb § tiʔəʔ sgʷəlub

tiləb=əxʷ ʔu–dxʷs–x%ʷuƛ’u–t–b–əb tiʔəʔ sgʷəlub

immediately=now PFV–CTD–chewed–ICS–PASS–DSD PROX pheasant

‘right away Pheasant was wanted to be chewed up (by the dogs)’

j. ləʔux%ʷ

lə=ʔux%ʷ Ø

PROG=go 3SUB

‘he is going along’

k. x%ʷul’ ləʔibəš

x%ʷul’ lə=ʔibəš Ø

only PROG=travel 3SUB

‘he is just walking around’

l. ʔudxʷsx%ʷuƛ’utəbəbəxʷ ʔə tiʔəʔ sqʷəbqʷəbayʔ

ʔu–dxʷs–x%ʷuƛ’u–t–b–əb=əxʷ ʔə tiʔəʔ sqʷəb–qʷəbayʔ

PFV–CTD–chewed–ICS–PASS–DSD=now PR PROX DSTR–dog

‘he was wanted to be chewed up by the dogs’

m. gʷəl huy ʔux%uƛ’utəbəxʷ

gʷəl huy ʔu–x%uƛ’u–t–b=əxʷ Ø

SCONJ SCONJ PFV–chewed–ICS–PASS=now 3SUB

‘and then he was chewed on’

n. ↑ gʷəl huy ləcutəbəxʷ, gʷiid ti adsqʷəbayʔ, sgʷəlub

gʷəl huy lə=cut–t–b=əxʷ gʷi–d ti ad–sqʷəbayʔ sgʷəlub

SCONJ SCONJ PROG=say–ICS–PASS=now call–ICS SPEC 2SG.PO–dog pheasant

‘and he was spoken to, Call your dogs, Pheasant!’

(Hess 1998: 78 – 79, lines 16 – 30)

539

The first line in (731) begins with a resetting of the F0 from a low at the end of the preceding

discourse episode, marking the line as the beginning of a new episode and its subject, sgʷəlub

‘Pheasant’, as the new discourse Topic which becomes the antecedent of the zero subjects in the

subsequent lines in the episode (with the exception of 731d, which seems to represent an

interjection on the part of the narrator). In the next line, (731e), the F0 is reset again as the scene

is set for the next episode, which describes the people (supernatural hunters and their dogs) that

Pheasant encounters in his wanderings. The next reset comes at line (731i), where the Topic once

again becomes Pheasant, the subject of the passive verb dxʷsx&ʷuƛ’utəbəb ‘be wanted to be

chewed’; this Topic is maintained over the next few lines until (731n), where F0 reset is used to

begin a new episode, recounting what the hunters said to Pheasant. As can be seen even in this

short excerpt, F0 reset at episode boundaries can be the sole marker of the line as a topic-setter

(as in 731a and i, which are morphosyntactically unmarked sentences), or it may combined with

a more specialized structure used as an overt marker of a shift in Topic or change of scene, as

with the non-verbal predicate construction in (731e). In fact, the interaction of F0 reset, the

setting and manipulation of discourse Topics, and the various morphosyntactic devices used in

Topic-setting and reference-tracking is an extremely complex one and depends profoundly on

things such as style, narrative skill, and culturally-bound conventions of storytelling, all of which

are issues well beyond the scope of this grammar.

11.3.2 Topical object marker

-ag„id

In conservative Lushootseed style, the topical object marker

-ag„id

([TOBJ]) is added to a

verb with a third-person object co-referent with the discourse topic (Kinkade 1990; Hess 1995).

Consider the text in (732):

540

(732) a. ʔukʷədad tiʔəʔ p’əč’əbulic’aʔs

ʔu–kʷəda–d Ø tiʔəʔ p’əč’əb•ulic’aʔ–s

PFV–take–ICS 3SUB PROX bobcat•blanket–3PO

‘he (Bobcat) took his bobcat-blanket’

b. gʷəl ʔald kʷədiʔ ʔadᶻalus

gʷəl ʔal–d Ø kʷədiʔ ʔadᶻalus

SCONJ at–ICS 3SUB REM:DMA beautiful

‘and he put it in a beautiful (place)’

c. gʷəl ləgʷədil ʔəxʷčəgʷ•us

gʷəl lə=gʷəd–il Ø ʔəs–dxʷ–čəgʷ•us

SCONJ PROG=sit–INCH 3SUB STAT–CTNRPT–seaward•face

‘and he sat facing the water’

d. diʔiɬ kʷi šudəgʷid ʔə tiʔəʔ č’ač’as

diʔiɬ kʷi s=šuɬ–d–əgʷid ʔə tiʔəʔ č’ač ’as

suddenly REM NM=see–ICS–TOBJ PR PROX child

‘all of a sudden the boy saw him’

(lit. ‘the sighting him of the boy [was] sudden’)

e. diɬəxʷ bayəʔ

diɬ=əxʷ bayəʔ Ø

FOC=now daddy 3SUB

‘that’s Daddy’

f. diɬəxʷ bayəʔ

diɬ=əxʷ bayəʔ Ø

FOC=now daddy 3SUB

‘that’s Daddy’

g. diɬ ti p’əč’əb ti ʔucutəb ʔə tiʔəʔ sqaqagʷəɬ

diɬ ti p’əč’əb ti ʔu–cut–t–əb Ø ʔə tiʔəʔ sqaqagʷəɬ

FOC SPEC bobcat SPEC PFV–speak–ICS–PASS 3SUB PR PROX noble.child

‘it was Bobcat whom the noble child spoke of’

(lit. ‘the one spoken of by the noble child [was] this one, Bobcat’)

(Hess 1995: 118)

This text occurs in a discourse episode where the Topic is Bobcat. Bobcat corresponds to the

zero third-person subject in every sentence in which he is an event-participant except in (732d).

Here, the object of the nominalized verb šudəgʷid ʔə tiʔəʔ č’ač’as ‘the child’s sighting of him’,

rather than the subject of the matrix clause, is co-referential with the topical Bobcat, and the verb

is consequently marked with the topical object marker. Because -agʷid is found only in the

541

speech of the oldest speakers, the exact nature of this marker, its syntax, and morphological

status are, at this stage, still somewhat uncertain, given its near-absence from the current corpus

of analyzed texts. As this is expanded to include more material from the eldest generation of

recorded speakers, it is to be hoped that more instances of -agʷid will be found and we will

come to a better understanding of this suffix and its discourse functions.

542

Appendix 1: Forms based on free radicals

√ʔa ‘be there, exist’

ʔaʔəd ‘put  there’

ʔaʔil ‘come to be in a place’

ʔatxʷ ‘put there’

ʔayid ‘put  there for ’

ʔaʔəb ‘be in a certain place’

√ʔalalus ‘happen’

ʔalalustxʷ ‘do to ’

√ʔaɬ ‘be fast’

ʔacut ‘hurry up’

√ʔəƛ’ ‘come’

ʔəƛ’c come after ’

ʔəƛ’txʷ ‘bring ’

√ʔibəš ‘travel, walk’

ʔibəštxʷ ‘make  travel’

√ʔigʷəɬa ‘climb tree’

ʔigʷəɬaac ‘climb after ’ (Sk.)

√ʔil ‘sing’

ʔilalikʷ ‘interpret ’

ʔilid ‘sing ’

ʔilyid ‘sing  for ’

√ʔistaʔ ‘be the same’

ʔistaʔtxʷ ‘do the same to ’

√ʔix&ʷ ‘be thrown; have thrown to’

ʔix&ʷid ‘throw  away’

√ʔukʷukʷ ‘play, have fun’

ʔukʷukʷbid ‘make fun of ’

√ʔuləx& ‘gather , forage for ’

ʔuləx&yid ‘gather  for ’

√ʔup’ ‘be seated on a lap’

ʔup’bid ‘sit on ’s lap’

ʔup’txʷ ‘sit a person on ’s lap; sit  on a person’s lap’

ʔup’ud ‘seat one’s lap’

√ʔux&ʷ ‘go’

ʔux&ʷc ‘go to ’

ʔux&ʷtxʷ ‘take ’

√bali ‘be forgetful’

baliic ‘forget about ’

√bap ‘be busy’

bapad ‘pester ’

√baqʷuʔ ‘be snow-covered’

baqʷuʔb ‘snow’

√bəč ‘be lying, be fallen from standing’

bəčad ‘set  down’

bəčagʷil ‘lie down’

bəčalikʷ ‘bet , place a wager with ’

dxʷbəčəb ‘sink, drown’

dxʷbəčəbagʷil ‘go under water’

dxʷbəčəbəd ‘sink ’

√bək’ʷ ‘all, completely’

bək’ʷdxʷ

‘manage to get all of ’

bək’ʷil ‘get used up, be done exhaustively’

bək’ʷildxʷ

‘consume of all of ’

√bəlx&ʷ ‘be beyond’

bəlx&ʷəd ‘pass ’

√buus ‘four’

buusil ‘become four’

√caq’ ‘be speared, be impaled’

caq’ad ‘spear ’

caq’alikʷ ‘spear , impale ’

caq’šadxʷ ‘lead ’

dxʷcaq’ax&adid ‘spear in the side’

√cikʷ ‘be straight, be taughtened’

ckʷab ‘be taut’

√cil ‘be dished up’

cilalikʷ ‘dish  (food)’

cildxʷyid ‘serve  to ’

cilid ‘dish  out’

cilyialikʷ ‘dish up  (food) for ’

cilyid ‘serve  to ’

√cukʷ ‘be unique’

cukʷtxʷ ‘make  the only one’

cugʷaɬ ‘be the last ’

543

√cut ‘speak’

cut ‘speak to ’

cuuc ‘speak to ’

dxʷcutəbid ‘catch on to ’

√c’aʔkʷ ‘be washed’

c’agʷačiʔb ‘wash hands’

c’agʷad ‘wash ’

√čaʔkʷ ‘seaward’

čagʷəb ‘be at sea’

čagʷəd ‘take  out to sea’

čagʷil ‘get out to sea’

čaʔkʷdxʷ ‘manage to get  to sea’

čaʔkʷtxʷ ‘take  out to sea’

dxʷčaʔkʷtxʷ ‘take  seaward’

√čac ‘be hidden’

čactxʷ ‘hide ’

čadᶻil ‘become hidden, hide self’

√čal

‘be overtaken’

čalad ‘chase ’

čaldxʷ ‘catch up to ’

čaltxʷ ‘catch ’

√čəbaʔ ‘be loaded down with ’

čəbaʔtxʷ ‘pack  on one’s back’

√čubə ‘go inland’

čubaac ‘go inland after ’

čubətxʷ ‘take  ashore’

√č’axʷ ‘be hit with a stick’

č’axʷad ‘club ’

č’axʷalikʷ ‘club ’

č’axʷdxʷ ‘manage to club ’

√č’əlp ‘turn’

č’əlpačiʔid ‘twist ’s wrist’

√č’it ‘nearby’

č’itcut ‘come close, approach’

č’itil ‘draw near’

č’itis ‘approach ’

√daʔ ‘be named’

daʔad ‘name ’

√dəkʷ ‘be inside’

dəgʷad ‘put  inside’

√diʔ ‘on the other side’

diʔil ‘go off a-ways’

√dukʷ ‘be anormal’

dukʷil ‘become strange, become supernatural’

dukʷildxʷ

‘be dissatisfied with ’

dukʷtxʷ ‘get angry with ’

dukʷud ‘change ’

√dᶻakʷ ‘be shaky, be shaking’

dᶻakʷaab ‘wag tail’

dᶻakʷad ‘rock ’

dᶻakʷtəd ‘rocking chair’

√dᶻal ‘turn around, turn over’

dᶻalqəd ‘turn  around’

dᶻalqʷusbid ‘look over shoulder at ’

dᶻəlax&adbid ‘visit ’

dᶻəlulčbid ‘turn towards ’

√dᶻaƛ’ ‘be confused’

dᶻaƛ’əd ‘confuse ’

dᶻaƛ’bid ‘be confused by , forget ’

dᶻaƛ’əb ‘get confused’

√dᶻəkʷ ‘travel, wander’

dᶻəkʷud ‘lead  astray, mislead ’

√dᶻix& ‘be broken down, be fallen apart’

dᶻix&id ‘break  down, take  apart’

√gəlk’ ‘be wound, be tangled’

gəlk’alikʷ ‘knit ’

gəlk’əd ‘wind ’

√gʷa ‘accompany, go along’

gʷahbid ‘accompany ’

gʷahtxʷ ‘take  along’

√gʷaagad ‘speak’

gʷaagatxʷ ‘speak to ’

√gʷax&ʷ ‘take a stroll’

gʷax&ʷtxʷ ‘take  for a walk’

√gʷəd ‘down’

gʷədil ‘sit down’

gʷədiltxʷ ‘sit  down’

gʷədiluɬ ‘go there to sit down’

544

gʷədis ‘sit down next to ’

√gʷəhaw’ə ‘apparently’

gʷəhaw’ad ‘blurt out ’

√gʷəx& ‘be untied’

gʷəx&ad ‘untie ’

gʷəx&agʷil ‘untie oneself’

√gʷi ‘make an invitation’

gʷiʔalikʷ ‘ask for ; seek compensation for ’

gʷiid ‘invite , call to ’

√haʔɬ ‘good’

haʔləb ‘be nice [weather]’

haʔɬid ‘make  good’

haʔɬil ‘become good’

√həd ‘be hot’

dxʷshəd ‘be hot in container; be hot (weather)’

dxʷshədil ‘heat up in container; get hot (weather)’

hədačiʔəb ‘warm up hands’

hədil ‘warm up’

√hədʔiw’ ‘be inside a house’

hədʔiw’b ‘go inside’

hədʔiw’c ‘go inside after ’

hədʔiw’dxʷ

‘manage to get inside’

√həliʔ ‘be alive’

həliʔdxʷ ‘save the life of ’

həliʔil ‘become well, heal’

həliʔis ‘live on ’

həliʔtxʷ ‘cure ’

√hiiɬ ‘be happy’

hiiɬbid ‘be happy about ’

√hikʷ ‘big’

hikʷtxʷ ‘respect ’

higʷəd ‘uphold , support ’

higʷil ‘become noble’

higʷild ‘make  bigger’

√hiqab ‘too much’

hiqabil ‘become too much’

√hud ‘burn’

hudčupyid ‘put  into the fire for ’

hudyid ‘make a fire for ’

√huy ‘be done, be made, be finished’

huyalikʷ ‘make , create ’

huydxʷ ‘manage to do ’

huydxʷyid ‘set up  for ’

huygʷastxʷ ‘marry ’

huyid ‘make  for ’

huyil ‘become’

huyud ‘make ’

√kiis ‘stand up’

kiistxʷ ‘stand  up’

√kʷatač ‘climb’

kʷatačtxʷ ‘carry  up tree’

√kʷəd ‘be held, be taken’

kʷədab ‘take  for self’

kʷədabid ‘take  captive’

kʷədabyid ‘make  captive’

kʷədačiʔalikʷ ‘shake hands with ’

kʷədad ‘take ’

kʷədalikʷ ‘take  over and over’

kʷədax&adid ‘take by the arm’

kʷədbid ‘steal from ’

kʷədxʷ

‘manage to take ’

kʷədyid ‘take  from ’

√kʷukʷcut ‘cook ’

kʷukʷcutyid ‘cook  for ’

√k’ʷəɬ ‘pour out, spill out’

k’ʷəɬdxʷ ‘spill ’

k’ʷɬalikʷ ‘serve  (liquid)’

√k’ʷič’ ‘butcher ’

k’ʷič’id ‘butcher ’

√k’ʷil ‘peek’

k’ʷilid ‘peek at ’

k’ʷilil ‘peer’

√k’ʷit’ ‘go down to shore’

k’ʷit’txʷ ‘take  down to shore’

545

√lab ‘appear’

labdxʷ ‘see ’

√laq ‘be last’

laqil ‘become last’

ləqbid ‘be behind ’

√lax& ‘recall, remember’

lax&bid ‘remember ’s story’

lax&c ‘think of ’

lax&dxʷ ‘remember ’

lax&txʷ ‘remind ’

√lək’ʷ ‘be eaten’

lək’ʷdxʷ

‘manage to eat ’

lək’ʷdxʷyid ‘manage to eat  away from ’

lək’ʷəd ‘eat ’

lək’ʷyid ‘eat  away from ’

√ləliʔ ‘be different’

ləliʔcut ‘change oneself’

ləliʔil ‘become different’

√ləq ‘listen’ (Sk)

ləqaladiʔbid ‘overhear ’

ləqc ‘listen to ’

√ləx& ‘be light’

ləx&il ‘get light, get bright’

ləx&šadid ‘light ’s way’

√lil ‘far’

lild ‘move away’

lildxʷ ‘draw away from ’

lis ‘go over to ’

√luƛ’ ‘be old’

luƛ’əb ‘age’

luƛ’il ‘grow old, grow up’

√ɬaʔ ‘arrive at a specific place’

ɬaʔačiʔbid ‘touch  with hand’

ɬaʔtxʷ ‘bring  to a place’

√ɬač’ ‘go out (fire)’

ɬač’alikʷ ‘fight fire’

ɬač’!d ‘put out  (fire)’

√ɬaq’ ‘be fallen, be lying down’

ɬaq’ad ‘put  down’

√ɬaw’t ‘be new’

ɬaw’tbid ‘be new for ’

√ɬax& ‘be dark’

ɬax&il ‘get dark’

√ɬəgʷɬ ‘leave ’

ɬəgʷəlbid ‘leave  behind, leave ’s presence’

ɬəgʷəldxʷ

‘manage to leave  behind’

ɬəgʷlyid ‘leave  for ’

√ɬič’ ‘get cut with knife’

ɬič’ib ‘cut  (cattails) for mats’

ɬič’ibtxʷ ‘bring  to cut cattails’

ɬič’id ‘slice ’

ɬič’šadid ‘amputate ’s leg’

√ɬid ‘be tied’

ɬidid ‘tie ’

ɬidtxʷ ‘tie to ’

ɬiʔɬda(hə)b ‘troll’

√ɬil ‘make a gift of food’

ɬild ‘give food to ’

ɬilyid ‘give  (food) to ’

√ƛ’al’ ‘also’

ƛ’al’yib ‘add ’

√ƛ’al ‘put  on’

ƛ’aldxʷ

‘manage to get  on’

ƛ’alšədid ‘put ’s shoes on him’

√ƛ’aɬəb ‘be salty’

ƛ’aɬəbalikʷ ‘salt ’

√ƛ’ax&ʷ ‘grow’

ƛ’ax&ʷtxʷ ‘bring up , raise ’

√ƛ’iq ‘emerge’

ƛ’iqagʷil ‘come out of’

ƛ’iqid ‘take  out from within’

√ƛ’iq’ ‘sticky’

ƛ’iq’ačiʔbtxʷ ‘make ’s hands sticky

ƛ’iq’ačiʔyibtxʷ ‘make ’s hands sticky’

√ƛ’ub ‘good, well’

ƛ’ubad

‘agree to ’

ƛ’ubil ‘get better’

ƛ’ubildxʷ

‘manage to improve ’

ƛ’ubtxʷ

‘have  fixed’

546

√pəd ‘be dirty, be buried’

dxʷpədəb ‘have dust or dirt inside’

pədičəd ‘dirty ’

√pus ‘be hit by  (missile)’

pusdxʷ ‘manage to hit  with ’

pusilyid ‘throw  for ’

pusud ‘throw at  with ’

√p’alil ‘revive’

p’alilcut ‘come to one’s senses’

√p’ayəq ‘carve canoe’

p’ayəqbid ‘hew , carve ’

√p’əq’ ‘drift’

dxʷp’əq’ ‘be filled with drifting things’

√p’il ‘be flat’

p’iləb ‘go flat’

√qada ‘steal ’

qadabid ‘steal ’

qadadid ‘steal  from ’

qadadyid ‘steal  for ’

√qa ‘be a lot’

qahil ‘become a lot’

√qəɬ ‘be awake’

qəɬdxʷ ‘accidentally awaken ’

√q’iƛ’ ‘be wounded’

q’iƛ’cut ‘injure oneself’

√qiq’ ‘be confined’

qiq’əd ‘confine ’

√qʷal ‘be marked, be painted’

dxʷqʷalusəb ‘paint face’

√qʷat ‘be lying; snow falls’

qʷatad ‘lay  out’

qʷatil ‘become laid out’

√qʷib ‘be ready’

qʷibid ‘prepare ’

√qʷiq’ʷ ‘be strong’

qʷiq’ʷil ‘become strong’

√qʷiʔad ‘yell’

qʷiʔaac ‘call out to ’

√qʷšab ‘be foggy’

qʷšabəd ‘fog  up’

√qʷuʔqʷa ‘have a drink’

qʷuʔqʷaʔiluɬ ‘go for a drink’

qʷuʔqʷad ‘drink  (liquid)’

qʷuʔqʷadid ‘drink ’

qʷuʔqʷadyid ‘drink  (drink) of ’s’

√q’axʷ ‘be frozen’

q’axʷad ‘freeze ’

√q’əlb ‘camp out’

q’əlbiluɬ ‘go camping’

√q’əls ‘cook with steam’

q’əlsəd ‘steam ’

√q’il ‘be aboard; ride in canoe’

q’ilagʷil ‘climb aboard’

q’ilb ‘put  on board one’s canoe’

q’ilid ‘put  on board’

q’iltxʷ ‘take  by canoe’

√q’p ‘form a lump; cramp up (muscle)’

q’pud ‘gather up ’

√q’ʷəl ‘be cooked, be ripe’

q’ʷəlb ‘cook  for oneself’

q’ʷəld ‘cook ’

q’ʷəlil ‘get hot’

q’ʷəlilc ‘warm stones to cook ’

√q’ʷib ‘be disembarked, be unloaded’

q’ʷibid ‘unload  from conveyance’

√q’ʷuʔ ‘be together with ’

q’ʷuʔbid ‘be together with ’

q’ʷuʔəd ‘gather ’

√saq’ʷ ‘fly’

saq’ʷtxʷ ‘fly off with ; fly  [airplane]’

√saʔ ‘be bad’

saʔil ‘get in trouble’

saʔtxʷ ‘dislike , hate ’

√sula ‘be in the middle’

sulabid ‘be in the middle relative to ’

sulayid ‘set  before ’

√šab ‘be dry’

dxʷšab ‘dry out (container or body of water)’

šabalikʷ ‘dry  (food)’

šabəb ‘dry out’

√šədᶻal ‘go outside’

šədᶻaldxʷ ‘manage to get outside’

šədᶻaltxʷ ‘take  outside’

√šəɬ ‘make ’

šəɬt’əbiɬədtxʷ ‘make  into rope’

√šidᶻ ‘launch sneak attack’

šidᶻəd ‘attack  by stealth’

547

√šq ‘high’

šəqəd ‘move  up high’

šəqlax&adəb ‘raise arms’

šqil ‘rise up’

√šub ‘disappear’

šubud ‘make  disappear; massacre ’

√šuk’ʷ ‘powder’

šuk’ʷil ‘turn grey’

√šul ‘be in, be under’

šulagʷil ‘enter cramped space’

šulud ‘pass underneath ’

√šuɬ ‘appear, be visible’

dxʷšəɬəb ‘look at  through water’

šudxʷ ‘catch sight of ’

šuɬbid ‘expect , keep an eye out for ’s arrival’

šuɬtxʷ ‘show to ’

šuuc ‘look at ’

√takʷ ‘be bought’

tagʷəlikʷ ‘buy ’

tagʷš ‘buy ’

√tay ‘go raiding’

tayc ‘come after  in raid’

√təč ‘roll off, tumble down’

təǰəd ‘roll ’

√təɬ ‘be true’

təɬtxʷ ‘make  true, speak truth’

√təlawil ‘run’

təlawiltxʷ ‘run off with ’

təlawis ‘run after ’

√təlč ‘be wide of mark’

təlčbid ‘miss  (throwing)’

√təɬəɬ

‘arrive safely’

təɬəɬdxʷ

‘manage to arrive safely’

√təxʷ ‘buy ’

təxʷtxʷyid ‘buy  for  ‘

√tq’ ‘slap’

tq’aʔɬdəlid ‘slap  in mouth’

√t’agʷt ‘be on top’

t’agʷil ‘get on top’

t’agʷtəd ‘put on top’

√t’aq’t ‘inland, landward’

dxʷt’aq’tcut ‘take oneself to higher ground’

dxʷt’aq’txʷ ‘take  ashore’

√t’əbaʔ ‘have fallen in water’

t’əbaʔagʷil ‘jump overboard’

√t’ilib ‘sing’

t’ilibtxʷ ‘sing to ’

√t’q ‘be thick’

t’qalikʷ ‘make bread; plaster’

√t’q’ ‘be patched (with stickum)’

t’q’abid ‘put stickum on ’

t’!q’əd ‘patch ’

√t’uc’ ‘be shot, fired on’

t’uc’ud ‘shoot  (target)’

√t’uk’ʷ ‘go home’

t’uk’ʷtxʷ ‘take  home’

√wač ‘keep watch’

wačbid ‘watch ’

√wəliʔ ‘be visible, be apparent’

wəliʔil ‘become visible’

√wiliq’ʷ ‘make an enquiry’

wiliq’ʷid ‘ask of ’

√xʷiʔ ‘no, not’

xʷəɬ ‘run out of ’

xʷiʔəd ‘refuse ’

xʷiʔil ‘become non-existent’

xʷiʔtxʷ ‘refuse to allow ’

√xʷiʔxʷiʔ ‘hunt for , forage for ’

xʷiʔxʷiʔiluɬ ‘go hunting/foraging’

√x&al ‘be written’

x&alad ‘write ’

x&altəd ‘writing implement’

√x&ayəb ‘laugh’

x&ayəbtxʷ ‘smile at ’

√x&əc ‘be afraid’

x&əcbid ‘fear ’

√x&əč ‘think, feel, use one’s mind’

x&čalikʷ ‘count ’

x&əčbid ‘intend ’

√x&əɬ ‘be sick’

x&əɬdxʷ ‘injure ’

√x&ilix& ‘be at war’

x&ilix&txʷ ‘make war on ’

√x&ix&iʔ ‘be ashamed’

x&ix&ibid ‘be ashamed of ’

√x&ix&q’ ‘insist’

x&ix&q’cut ‘talk back, argue’

√x&ʷal’ ‘be unable, fail, lose’

x&ʷal’bid ‘be unable to manage ’ ,

x&ʷal’dxʷ ‘get the better of ’

David Beck 10-2-7 2:19 PM

Comment: check to make sure there aren’t

BOTH an unaccusative and an unergative use

548

√x&ʷaq’ʷ ‘be worried, be preoccupied’

dxʷx&ʷax&ʷaq’ʷacut ‘be troubled’

x&ʷaq’ʷad ‘be concerned about ’

√x&ʷax&ʷaq’ʷ ‘feel concern’

x&ʷax&ʷaq’ʷcut ‘be troubled, be worried’

√x&ʷəc ‘be sharp’

dxʷx&ʷədᶻəb ‘be tart, be strong (coffee)’

√xʷəxʷaʔxʷaʔ ‘be lightweight’

xʷəxʷaʔxʷaʔəd ‘make  lighter’

√x&ʷil’ ‘be lost’

x&ʷil’alcbid ‘lose ’

√yabuk’ʷ ‘(to) fight’

yabuk’ʷbid ‘fight over ’

√yayus ‘do work’

yayusbid ‘work on ’

√yəhaw’ ‘only if; not until’

yəhaw’txʷ ‘go ahead with ’

√yəy’duʔ ‘swing in a swing’

yəy’duʔiluɬ ‘go for a swing’

√yiq’ ‘be worked into tight place’

yiq’ib ‘make  (baskets)’

yiq’id ‘weave  (basket)’

549

Appendix 2: Forms based on bound radicals

º√ʔab ‘be extended’

ʔabačiʔb ‘extend hands’

ʔabalikʷ ‘give  away as in a potlatch’

ʔabəd ‘extend , give ’

ʔabšədəb ‘extend legs’

ʔabucidid ‘bring  lunch’

ʔabyid ‘give to ’

º√ʔac ‘centre’

ʔacigʷədil ‘be inside of’

º√ʔadᶻq ‘be met’

ʔadᶻqbid ‘meet ’

ʔadᶻqdxʷ ‘happen to meet ’

º√ʔaladᶻ ‘care for ’

ʔaladᶻiʔlyid ‘babysit  for ’

º√ʔay’ ‘be traded’

ʔay’waʔsəd ‘exchange ’

º√ʔəɬ ‘eat’

ʔəɬdiluɬ ‘go out to eat’

ʔəɬdiluɬbid ‘go to eat off of ’

ʔəɬəd ‘feed on ’

181

ʔəɬtxʷ ‘feed ’

º√ʔəq’ʷ ‘be open’

dxʷʔəq’yax&adid ‘open  (door)’

º√ʔəy’ ‘be found’

ʔəy’cut ‘find oneself in a place’

ʔəy’dxʷ ‘find something’

º√ʔiʔab ‘be wealthy’

sʔiʔab ‘noble person’

ʔiʔabil ‘become wealthy’

º√ʔudəgʷ ‘centre’

ʔudəgʷičil ‘get into middle of house’

º√ʔul ‘sing’

ʔuliʔɬ ‘sing lullaby’

ʔulud ‘sing to ’

º√ʔuq’ʷ ‘be unplugged’

ʔuq’ʷdxʷ

‘be left open to , be vulnerable to ’

ʔuq’ʷud ‘unplug ’

º√bəq’ ‘have in mouth’

bəq’dxʷ ‘taste ’

bəq’əd ‘swallow ’

º√biƛ’ ‘be smashed, be crumbled’

biƛ’id ‘smash ’

biƛ’il ‘get crushed’

º√biqʷ ‘be loose’

biqʷid ‘loosen ; permit ’

biqʷyid ‘permit  to ’

º√c’adᶻax& ‘feel annoyance’

c’adᶻax&bid ‘be bothered by , be fed up with ’

c’adᶻaxtx&ʷ ‘bother ’

º√c’əb ‘clear land’

c’əbəb ‘pick [berry]’

c’əbəbiluɬ ‘go berry-picking’

º√c’əl ‘be defeated’

c’əlalikʷ ‘defeat ’

c’əld ‘defeat ’

c’əldxʷ

‘manage to defeat ’

º√c’əs ‘be pecked’

c’əsəd ‘peck ; nail ’

c’salikʷ ‘peck at ; nail ’

º√c’ic’əyikʷ ‘wink’

c’ic’əyikʷalusbid ‘wink at ’

º√c’ix& ‘be fried’

c’ix&alikʷ ‘fry ’

c’ix&id ‘fry ’

º√c’uqʷ ‘be sucked on’

c’uqʷačiʔb ‘suck on hand’

c’uqʷud ‘suck on ’

√čaɬaʔ ‘not be recognized’

čaɬaʔcut ‘distance oneself from ’

181

This stem is bivalent intransitive; the [-əd] in this form is not synchronically the internal causative suffix.

550

čaɬaʔdxʷ ‘not recognize ’

º√čc ‘be red’

čcil ‘turn red’

dxʷčəcəb ‘be red (water)’

x&ičəc ‘red’

º√č’aʔ ‘be dug up’

č’aʔalikʷ ‘dig for  (edible roots)’

č’aʔəb ‘dig for  (roots)’

č’aʔəd ‘dig for  (roots)’

º√č’ədᶻ ‘be stalked’

č’ədᶻalikʷ ‘stalk  (prey)’

č’ədᶻəd ‘sneak up on ’

º√dikʷ ‘be advised’

digʷid ‘advise ’

dikʷdxʷ

‘instruct ’

dxʷdigʷid ‘advise ’

xʷdikʷ ‘advice’

º√dᶻaq ‘be in mourning’

dᶻaqad ‘mourn ’

dᶻaqəbid ‘mourn for ’

º√dᶻub ‘be kicked’

dᶻubalikʷ ‘dance’

dᶻubalikʷiluɬ ‘go to a dance’

dᶻubud ‘kick ’

º√gəq ‘shining’

gəqəb ‘[sun] shines’

gəqil ‘become dazzling; clear up [weather]’

º√gʷəč’ ‘be sought’

gʷəč’alikʷ ‘habitually seek ’

gʷəč’əb ‘seek  for self’’

gʷəč’əd ‘look for ’

º√gʷəlal ‘be hurt’

gʷəlalalikʷ ‘kill , slaughter ’

gʷəlald ‘punish , kill ’

º√gʷu ‘bark (dog)’

gʷuhəb ‘bark’

gʷuhud ‘bark at ’

º√hay ‘be known’

hayəd ‘pay attention to ’

haydxʷ ‘know ’

haydxʷyid ‘find out about  for ’

º√hiq ‘be pushed’

hiqid ‘push ’

hiqil ‘get pushed up’

º√ǰuʔ ‘be glad’

ǰuʔid ‘entertain ’

ǰuʔil ‘have a good time, be happy’

ǰuʔilbid ‘be happy for ’

º√kʷaxʷ ‘be helped’

kʷaxʷad ‘help ’

kʷaxʷdxʷ

‘manage to help ’

º√kʷaʔ ‘be released’

kʷaʔd ‘let go of ’

kʷaʔdxʷ ‘manage to let go of ’

º√k’aw ‘be chewed’

k’awad ‘chew’

k’awalikʷ ‘chew ’

º√k’aw ‘be bumped’

k’awdxʷ

‘collide with ’

k’awqid ‘bump head’

º√k’ʷaɬ ‘look closely, peer’

k’ʷaɬad ‘examine ’

k’ʷək’ʷaɬəb ‘be near-sighted’

º√k’ʷalč’ ‘be bent backwards’

k’ʷalč’əb ‘bend self backwards’

k’ʷalč’əd ‘bend  backwards’

º√k’ʷəƛ’ ‘miss’

k’ʷəƛ’c ‘miss  (target)’

k’ʷəƛ’gʷasbid ‘miss meeting’

º√laʔ ‘be located’

laʔəd ‘locate ’

laʔyid ‘locate  for ’

º√ləc’ ‘have come down on’

ləc’əd ‘step on ’

ləc’yid ‘step on  affecting ’

º√ləxʷ ‘be stabbed, be cut’

ləxʷdxʷ ‘manage to stab ’

551

ləxʷud ‘stab ’

º√lu ‘hear’

ludxʷ ‘happen to hear , hear about ’

luhəladiʔ ‘hear ’

luuc ‘listen to ’

luud ‘hear ’

º√ɬal ‘be removed from fire’

ɬalil ‘get out of fire’

ɬalš ‘remove  from fire’

º√ɬuq’ʷ ‘be peeled’

ɬuq’ʷač ‘be bald (lit. ‘peeled head’)’

ɬuq’ʷud ‘peel ’

º√ƛ’ac’ ‘cinched up’

ƛ’ac’ahəb ‘cinch up belt’

º√ƛ’akʷ ‘be stitched’

ƛ’agʷəb ‘make  (mat)’

ƛ’agʷəd ‘stitch  (mat)’

º√ƛ’ip’ ‘be compressed’

dxʷƛ’ip’alikʷ ‘player in hand game’

ƛ’ip’id ‘squeeze ’

º√ƛ’uq’ʷ ‘be crammed in’

ƛ’uq’ʷagʷil ‘cram self into small space’

ƛ’uq’ʷud ‘plug  in’

º√pkʷ ‘be broken off’

pqʷud ‘break off ’

pqʷyid ‘break off  for ’

º√p’ic’ ‘be wrung out’

p’ic’albixʷ ‘milk (cow)’

p’ic’id ‘wring  out’

º√p’t’ ‘be stored’

p’t’ad ‘store , tidy ’

p’t’alikʷ ‘save ’

º√puʔ ‘be blown on’

dxʷpuhigʷədid ‘blow on ’

puʔud ‘blow on , blow  out’

º√qəd ‘fornicate’

dxʷqədid ‘cuckhold ’

dxʷqədid ‘have sex with  (spouse)of ’

qədəb ‘have illicit sex with ’

qədəbtxʷ ‘seduce  into adultery’

º√qʷc ‘slide, slip’

qʷcab ‘slip’

qʷcagʷil ‘slide down’

qʷcagʷis ‘slide down after ’

qʷəcəd ‘slide ’

º√qʷq’ʷ ‘be white’

qʷəq’ʷil ‘turn white’

x&iqʷəq’ʷ ‘white’

º√q’al ‘be fooled, be convinced’

q’alad ‘fool ’

q’albid ‘be fooled by ’

º√q’p’ ‘be compensated’

q’p’alikʷtxʷ ‘pay for a crime’

q’p’ud ‘pay ’

º√q’it’ ‘be hung’

q’it’id ‘hang ’

q’italikʷ ‘hang  (fish) up to dry’

º√sub ‘have odour’

subalikʷ ‘smell ’

subud ‘smell ’

º√suxʷt ‘be recognized’

suxʷtəbid ‘recognize ’

suxʷtəš ‘recognize ’

º√taʔ ‘be in place’

taʔəd ‘take  over there’

ʔəstaʔtxʷ ‘have  in place’

º√tatab ‘speak’

tatabəd ‘confer about ’

tatabtxʷ ‘talk to ’

º√təq’ ‘be slapped’

təq’aladiʔəb ‘slap sides of head’

º√tq ‘be closed’

təqdxʷ ‘block ’s path’

tqad ‘close , block  off’

º√ts ‘be punched’

təsəd ‘punch ’

tsalikʷ ‘hammer , pound ’

º√tudəq ‘be enslaved’

studəq ‘slave’

552

tudəqil ‘become enslaved’

º√tul ‘be interpreted’

tulalikʷ ‘read ; interpret ’

tulud ‘interpret ’

º√tup ‘be pounded’

tupud ‘pound ’

tupyib ‘pound to prepare as food’

º√tux&ʷ ‘be stretched’

tux&ʷud ‘stretch ’

tux&ʷalikʷ ‘stretch ’

xʷtux&ʷtux&alikʷ ‘(spider) stretches  (web)’

º√tx&ʷ ‘be pulled’

təx&ʷgʷiɬ ‘pull canoe’

tx&ʷud ‘pull on ’

º√t’iq’ʷ ‘be smoky/murky’

dxʷt’iq’ʷəb ‘be murky (water)’

dxʷt’iq’ʷil ‘get murky (water)’

t’iq’ʷalikʷ ‘smoke  (meat)’

º√t’iwiɬ ‘practice religion’

st’iwiɬ ‘religion’

t’iwiɬiluɬ ‘go to church’

º√t’ukʷ ‘be measured’

t’ugʷud ‘figure  out’

t’ukʷtəd ‘tape measure’

º√xʷac ‘be hoisted’

xʷacad ‘carry ’

xʷaʔxʷcəb ‘be rock-lifting’

º√xʷəb ‘be thrown’

xʷəbagʷil ‘throw oneself’

xʷəbaladiʔb ‘toss head side to side’

xʷəbaličbid ‘toss  (pack) onto own back’

xʷəbəd ‘toss ’

º√xʷš ‘be thrown’

xʷəšəd ‘throw ; give away ’

xʷšalikʷ ‘sow ; give goods at potlatch’

º√xʷt’ ‘be fallen, be descended’

xʷit’il ‘descend’

xʷit’ild ‘lower ’

xʷt’ad ‘take  down’

xʷt’agʷil ‘climb down’

º√xʷuyub ‘be sold’

xʷuyubtxʷ ‘sell ’

xʷuyubtxyidʷ ‘sell  for ’

º√x&əd ‘be pressed’

x&ədačiʔb ‘push hands away’

x&ədəd ‘push ’

x&ədyid ‘set  aside for ’

º√x&ƛ’ ‘be bitten’

x&ƛ’alikʷ ‘bite into ’

x&əƛ’əd ‘bite ’

º√x&ib ‘be grabbed, clawed’

x&ibid ‘grab  by throat’

x&ibigʷs ‘have one’s things in hand’

º√x&id ‘be growling’

x&idib ‘growl’

x&idid ‘growl at ’

º√x&q ‘be wrapped, be tied’

x&əqəd ‘wrap ’

x&əqšadəb ‘wrap legs’

x&əqšadid ‘bind legs of ’

x&qagʷil ‘tie oneself down’

x&qičyid ‘bind  into a pack for ’

º√x&ʷadᶻ ‘be injured’

x&ʷadᶻad ‘punish , annihilate ’

x&ʷadᶻalikʷ ‘slaughter ’

º√yaƛ’ ‘be dry’

yaƛ’ad ‘scoop up  (water)’

yaƛ’əb ‘carrywater’

º√yəc ‘report on ’

yəcbid ‘tell about ’

yəcəb ‘report on ’

yəcəd ‘report ’

º√yəhub ‘tell legend’

syəhub ‘myth, legend’

yəhubtxʷ ‘recite legend for ’

553

º√yəl ‘pair’

yəlačiʔbid ‘use both hands on ’

554

Appendix 3: Forms based on hypothetical radicals

*√ʔəy’ ‘be found’

√ʔəy’dxʷ ‘find ’

ʔəy’dxʷyid ‘find  for ’

*√ʔus ‘head’

√ʔusil ‘dive’

ʔusis ‘dive after ’

*√ʔuš ‘pity’

º√ʔušəb ‘feel pity’

ʔušəbid ‘feel pity for ’

*√bak’ʷ ‘move quickly’

√bak’ʷad ‘move  quickly’

*√baɬ ‘be cured by shamanism’

√baɬad ‘cure  with shamanism’

*√biq’ ‘be pressed down’

√biq’id ‘press ’

*√bis ‘be selected’

√bisəd ‘select ’

*√c’ip’ ‘be shut (eyes)’

√c’ip’lil ‘shut eyes’

c’ip’əlis ‘ignore ’

c’ip’lilbid ‘shut eyes to avoid seeing ’

*√čədᶻq’ʷ ‘be rubbed together’

√čədᶻq’ʷəd ‘rub  together’

*√čətx&ʷ ‘be gobbled up’

√čətx&ʷəd ‘gobble  up’

*√dᶻili ‘be despised’

√dᶻilid ‘despise ’

*√gəq’ ‘be opened’

√gəq’əd ‘open something’

*√gʷəc ‘wade’

√gʷəcil ‘wade’

gʷəcis ‘wade after ’

*√hil ‘obey’

√hilid ‘command ’

*√hiq’ʷ ‘be coveted’

√hiq’ʷəb ‘covet ,lust after ’

hiq’ʷəbid ‘lust after ’

hiq’ʷəbyid ‘covet  from ’

*√hiw ‘be advanced’

√hiwil ‘proceed’

hiwiltxʷ ‘go ahead with ’

hiwis ‘approach , go after ’

*√ɬal ‘be ashore’

√ɬalil ‘go ashore’

ɬalildxʷ ‘manage to get ashore’

ɬaliltxʷ ‘bring  ashore’

ɬalis ‘go ashore after ’

*√ɬal ‘removed from fire’

√ɬalš ‘remove  from fire’

*√ɬč ‘be on the way’

√ɬčil ‘arrive’

ɬčildxʷ ‘manage to arrive’

ɬčiltxʷ ‘arrive with ’

ɬčiltxʷyid ‘bring  for ’

ɬčilyid ‘arrive with  for ’

ɬčis ‘arrive at ’

ɬčisbid ‘visit and inconvenience them’

*√ƛ’aq’ʷ ‘be ambushed’

√ƛ’aq’ʷad ‘lie in wait for ’

*√pač ‘be laid out’

√pačad ‘lay  out’

*√paq’ ‘be scattered, distributed’

√paq’ad ‘distribute ’

*√pit ‘be aware’

√pitəb ‘pay attention, be aware’

pitəbid ‘pay attention to ’

*√q’x& ‘be insulted’

√q’x&ad ‘insult ’

*√qəl ‘be turbid’

√qələb ‘(to) rain’

dxʷqələbil ‘be turbid (water)’

*√qʷaɬ ‘be driven off’

√qʷaɬad ‘drive  (animal); drive  off’

555

*√saxʷ ‘jump, run’

√saxʷəb ‘jump, sprint’

saxʷəbid ‘run after  or up to ’

saxʷəbtxʷ ‘run off with , kidnap ’

*√sət’ ‘be raised’

√sət’əd ‘lift ’

*√sax& ‘be scraped’

√sax&ad ‘scrape ’

*√t’ič ‘wade’

√t’ičib ‘wade’

t’ičibtxʷ ‘make  wade’

*√təb ‘have a craving’

√təbaš ‘crave ’

*√tədᶻ ‘be in bed’

√tədᶻil ‘go to bed’

tədᶻiltxʷ ‘put  to bed’

tədᶻis ‘go to bed with ’

*√tudᶻ ‘be bent forward’

√tudᶻil ‘bend forward’

tudᶻis ‘bend over to get ’

*√tul ‘cross river’

√tulil ‘cross river’

tuliltxʷ ‘take  across river’

*√xʷak’ʷ ‘be tired’

√xʷak’ʷil ‘be tired’

xʷak’ʷilbid ‘become disaffected with ’

xʷak’ʷis ‘get tired of ’

*√xʷalus ‘be waved’

√xʷalusəd ‘wave ’

*√x&aƛ’ ‘argue’

√x&aƛ’il ‘argue’

x&aƛ’is ‘defend from ’

*√x&ay ‘laugh’

√x&ayəb ‘laugh’

x&ayəbid ‘laugh at ’

*√x&ʷub ‘be quiet’

√x&ʷubil ‘be quiet’

x&ʷubis ‘be quiet about ’

*√x&ʷuƛ’ ‘be chewed up’

√x&ʷuƛ’ud ‘chew  up’

556

Appendix 4: Texts included in the Lushootseed Corpus

bibščəb ʔi tiʔiɬ suʔsuq’ʷaʔs, tətyika ‘Young Mink and Tutyeeka’ as told by Edward Sam to T.

M. Hess at Tulalip in the summer of 1963 (published in Hess 1995)

sčətxʷəd ʔi tsiʔiɬ ƛ’aƛ’ac’apəd ‘Black Bear and Ant’ as told by Edward Sam to T. M. Hess at

Tulalip in the summer of 1963 (published in Hess 1995)

sbiaw ʔi tiʔiɬ hikʷ č’ƛ’aʔ ‘Coyote and the Big Rock’ as told by Edward Sam to T. M. Hess at

Tulalip in the summer of 1963 (published in Hess 1995)

sčətxʷəd ʔi tiʔiɬ c’ix&c’ix& ‘Black Bear and Fish Hawk’ as told by Edward Sam to T. M. Hess at

Tulalip in the summer of 1963 (published in Hess 1995)

ʔəsx&əɬ tsiʔəʔ k’aʔk’aʔ ‘Crow was sick I’ as told by Martha Lamont to Leon Metcalf at Tulalip in

1953 (published in Hess 1998)

ʔəsx&əɬ tsiʔəʔ k’aʔk’aʔ ‘Crow is sick II’ as told by Martha Lamont to T. M. Hess at Tulalip in the

summer of 1963 (published in Hess 1998; Bierwert 1993)

dukʷibəɬ ‘Changer’ as told by Martha Lamont to T. M. Hess at Tulalip in the summer of 1963

(published in Hess 1998; Bierwert 1993)

sgʷəlub ʔi tiʔəʔ qaw’qs ‘Pheasant and Raven’ as told by Martha Lamont to T. M. Hess at Tulalip

in the summer of 1963 (published in Hess 1998)

sbiaw ʔi tsiʔəʔ bədaʔ ‘Coyote and his daughter’ as told by Martha Lamont to T. M. Hess at

Tulalip in the summer of 1963 (published in Hess 1998)

ʔəslaɬlil tiʔiɬ təkʷtəkʷəlus ‘Owl lived there’ as told by Martha Lamont to T. M. Hess at Tulalip

in the summer of 1963 (published in Hess 2006)

čəgʷas ʔə tiʔəʔ sbəq’ʷaʔ tsiʔəʔ x&ʷuʔx&ʷəyʔ ‘Little Diver was the wife of Heron’ as told by Martha

Lamont to T. M. Hess at Tulalip in the summer of 1963 (published in Hess 2006)

ʔəbsčəgʷas tiʔəʔ bədaʔ ʔə tiʔəʔ sbiaw ʔə tə səsaʔliʔ ‘Coyote’s son had two wives’ as told by

Martha Lamont to T. M. Hess at Tulalip in the summer of 1963 (published in Hess 2006)

ʔəslaɬlil tiʔiɬ ʔi sgʷəlub ʔi tiʔiɬ x&əɬx&əɬtəds ‘Pheasant and his wife’s brothers lived there’ as told

by Martha Lamont to T. M. Hess at Tulalip in the summer of 1963 (published in Hess 2006;

Bierwert 1993)

Lady Louse as told by Elizabeth Krise to T. M. Hess

Star Child as told by Mary Willup to Leon Metcalf

Star Child as told by Harry Moses to Leon Metcalf

Star Child as told by Dora Salomon to Vi Taq’š$blu Hilbert

Mink and Tutyika I as told by Martha Lamont to Leon Metcalf

Mink and Tutyika II as told by Martha Lamont to T. M. Hess

Basket Ogress as told by Dewey Mitchell to Leon Metcalf

Basket Ogress as told by Alice Williams to Leon Metcalf

Basket Ogress as told by Julie Siddle to Leon Metcalf

Basket Ogress as told by Martha Lamont to T. M. Hess

Basket Ogress as told by Agnes James to Leon Metcalf

Basket Ogress as told by Louise Anderson to Leon Metcalf

557

Basket Ogress as told by Martin Sampson to Leon Metcalf

Stealing daylight as told by Harry Moses to Leon Metcalf (published as “How Daylight was

stolen” in Hilbert & Hess 1977)

558

Bibliography

The following bibliography includes, in addition to the references cited in the text, a fairly

complete listing of all the major linguistic and documentary work dedicated specifically to

Lushootseed. A more comprehensive bibliography of Lushootseed materials, including

anthropological as well as linguistic work, can be found in van Eijk (n.d.), on which I have relied

heavily in the compilation of this list.

Ballard, Arthur C. (1927). Some tales of the Southern Puget Sound Salish. University of

Washington Publications in Anthropology 2(3), 57–81.

Ballard, Arthur C. (1929). Mythology of Puget Sound. University of Washington Publications in

Anthropology 3(2), 31–150.

Ballard, Arthur C. (1935). Southern Puget Sound Salish kinship terms. American Anthropologist

37, 111–116.

Ballard, Arthur C. (1950). Calendric terms of the Southern Puget Sound Salish. Southwestern

Journal of Anthropology 6, 79–99.

Barthmaier, Paul. (1998). Lushootseed vowels, a preliminary phonetic study. 33

International

Conference on Salish and Neighbouring Languages 33, 1–11. Seattle, Washington.

Barthmaier, Paul. (2000). Lushootseed argument structure and the discourse function of the

morpheme /–b/. In Suzanne Gessner and S. Oh (eds.), Papers for ICSNL XXXV: The Thirty

Fifth International Conference on Salish and Neighboring Languages, 1–18. Vancouver:

University of British Columbia Working Papers in Linguistics.

Bates, Dawn. (1986). An analysis of Lushootseed diminutive reduplication. In Vassiliki

Nikiforidou, Mary VanClay, Mary Niepokuj and Deborah Feder (eds.), Proceedings of the

559

Twelfth Annual Meeting of the Berkely Linguistics Society, 1–13. Berkeley: University of

California Press.

Bates, Dawn. (1997a). Person marking in Lushootseed subordinate clauses. International Journal

of American Linguistics 63, 316–333.

Bates, Dawn. (1997b). Semantic Roles and Referent Tracking in Martha Lamont’s “Pheasant and

Raven.” 32

International Conference on Salish and Neighbouring Languages, 1–21. Port

Angeles, Washington.

Bates, Dawn. (1999). Distance in narrative time and space, aspect markers and determiner choice

in Martha Lamont’s “Pheasant and Raven.” 34

International Conference on Salish and

Neighbouring Languages, 1–11. Kamloops, B.C.

Bates, Dawn. (2002). Narrative functions of past tense marking in a Lushootseed text. In Carrie

Gillon, N. Sawai, and Rachel Wojdak (eds.), Papers for ICSNL XXXVII: The Thirty Seventh

International Conference on Salish and Neighboring Languages, 17–34. Vancouver:

University of British Columbia Working Papers in Linguistics.

Bates, Dawn. (2005). Code–switching in Martha Lamont’s “Little Diver was the wife of Heron”.

In J. C. Brown, Masaru Kiyota, and Tyler Peterson (eds.), Papers for ICSNL XL: The

Fortieth International Conference on Salish and Neighbouring Languages, 32–48.

Vancouver: University of British Columbia Working Papers in Linguistics.

Bates, Dawn, and Thomas M. Hess. (2001). Tense or aspect? A prefix of future time in

Lushootseed. In Leora Bar-El, Linda T. Watt, and Ian Wilson (eds.), Papers for ICSNL

XXVI: The Thirty-sixth International Conference on Salish and Neighbouring Languages,

25–36. Vancouver: University of British Columbia Working Papers in Linguistics.

560

Bates, Dawn, and Thomas M. Hess. (2004). The expression of NPs in Lushootseed Texts. In J.

C. Brown and Tyler Peterson (eds.), Papers for ICSNL XXIX: The Thirty-ninth International

Conference on Salish and Neighbouring Languages. University of British Columbia

Working Papers in Linguistics 14, 1–44. Vancouver: University of British Columbia

Working Papers in Linguistics.

Bates, Dawn and Thomas M. Hess. (2003). An agentive suffix in Lushootseed. In J.C. Brown

and Michele Kalmar (eds.), Papers for ICSNL XXXVIII: The Thirty Eighth International

Conference on Salish and Neighboring Languages, 1–6. Vancouver: University of British

Columbia Working Papers in Linguistics.

Bates, Dawn, Thomas M. Hess, and Vi Hilbert. (1994). Lushootseed dictionary. Seattle:

Washington University Press.

Beck, David. (1996). Transitivity and causation in Lushootseed morphology. Canadian Journal

of Linguistics 41, 109–140.

Beck, David. (1997). Theme, Rheme, and communicative structure in Lushootseed and Bella

Coola. In Leo Wanner (ed.), Recent trends in Meaning-Text Theory, 93 – 135. Amsterdam:

John Benjamins.

Beck, David. (1999). Words and prosodic phrasing in Lushootseed narrative. In Toby A. Hall

and Ursula Kleinhenz (eds.), Studies on the Phonological Word, 23–46. Amsterdam: John

Benjamins.

Beck, David. (2000a). Nominalization as complementation in Bella Coola and Lushootseed. In

Kaoru Horie (ed.), Complementation: Cognitive and Functional Perspectives, 121–147.

Amsterdam: John Benjamins.

561

Beck, David. (2000b). Semantic agents, syntactic subjects, and discourse topics: How to locate

Lushootseed sentences in space and time. Studies in Language 24, 277–317.

Beck, David. (2009). A taxonomy and typology of Lushootseed valency–increasing suffixes.

International Journal of American Linguistics 75, 533–569.

(2009). Thematicity in Lushootseed syntax. In David Beck, Kim Gerdes, Jasmina Milićević, and

Alain Polguère (eds.), Proceedings of the Fourth International Conference on Meaning-Text

Theory, 55–64. Montreal: OLST.

(to appear). Uni-directional flexibility and the noun–verb distinction in Lushootseed. In Jan

Rijkhoff and Eva van Lier (eds.), Flexible word classes: A typological study of

underspecified parts-of-speech. Amsterdam: Benjamins.

(to appear). Communicative Structure in Lushootseed syntax: Thematicity and Focalization. In

José Camacho, Rodrigo Gutiérrez-Bravo, and Liliana Sánchez (eds.), Information structure

in languages of the Americas. Berlin: Mouton de Gruyter.

Beck, David and David Bennett. (2007). Extending the Prosodic Hierarchy: Evidence from

Lushootseed narrative. Northwest Journal of Linguistics 1: 1–34. (http://www.sfu.ca/nwjl/)

Bennett, David, and David Beck. (1998). Extending the Prosodic Hierarchy: Evidence from

Lushootseed narrative. 33

International Conference on Salish and Neighbouring

Languages, 35–59. Seattle, Washington.

Bianco, Violet M. (1995). Stress in Lushootseed: A preliminary analysis. 30

International

Conference on Salish and Neighbouring Languages. Victoria, B.C.

Bierwert, Crisca. (1993). “Poetic Fancy”, a glimpse at the translative commentary of Martin J.

Sampson. In Arnold Krupat (ed.), New voices in Native American literary criticism, 529–

542. Washington, D. C.: Smithsonian Institution.

562

Bierwert, Crisca (1996). Lushootseed texts: An introduction to Puget Sound Salish narrative

aesthetics. Lincoln, NB: University of Nebraska Press.

Chirouse, Eugene C. (1880). A short method to learn the Snohomish Indian Language in 14

Lessons. Manuscript No. HHB [P–B86] in the Bancroft Library, University of California,

Berkeley.

Cook, Suzanne Elizabeth. (1999). Rhetorical structure of a Lushootseed (Salish) narrative. MA

thesis, University of Victoria.

Czaykowska-Higgins, Ewa and Kinkade, M. Dale (1998). Salish languages and linguistics. In

Ewa Czaykowska-Higgins and M. Dale Kinkade (eds.), Salish languages and linguistics:

Theoretical and descriptive perspectives, 1–68. Berlin: Mouton de Gruyter.

Davis, Henry. (2000). Remarks on Proto-Salish subject inflection. International Journal of

American Linguistics 66, 499–520.

Davis, Henry. (2005). On the syntax and semantics of negation in Salish. International Journal

of American Linguistics 71, 1–55.

Davis, Philip W., and Saunders, Ross. (1978). Bella Coola syntax. In Eung-Do Cook and

Johnathan Kaye (eds.), Linguistic studies of native Canada, 37–65. Vancouver: University

of British Columbia Press

Dryer, Matthew S. (1986). Primary objects, secondary objects, and anti-dative. Language 62,

808–845.

Galin, Anne. (1982). The encoding of spatial relations in Lushootseed. 17

International

Conference on Salish and Neighbouring Languages, 67–88. Portland, Oregon.

Galin, Anne. (1983). Spatial organization in Lushootseed culture, texts and language. Ph.D.

dissertation, Columbia University.

563

Galin, Anne. (1984). Cognitive Configurations in Lushootseed. 17

International Conference on

Salish and Neighbouring Languages, 66–80. Victoria, B.C.: Working Papers of the

Linguistics Circle of the University of Victoria.

Foley, William, and Robert D. van Valin, Jr. (1985). Information packaging in the clause. In

Timothy Shopen (ed), Language Typology and Syntactic Description, 282–364. Cambridge:

Cambridge University Press.

Gerdts, Donna B. (1988). Object and absolutive in Halkomelem Salish. New York: Garland.

Gerdts, Donna B. (2006). Argument Realization in Halkomelem: A Study in Verb Classification.

In Atsushi Fujimori and Maria Amelia Reis Silva (eds.), Proceedings of the 11

Workshop

on Structure and Constituency in the Languages of the Americas, 61–81. Vancouver:

University of British Columbia Working Papers in Linguistics.

Gibbs, George. (1891). Dictionary of the Nisqually. In John W. Powell, Indian linguistic families

of America North of Mexico. Bureau of American Ethnology Annual Report 7, 285–361.

Washington, D.C.: Government Printing Office.

Haeberlin, Herman. (n.d.). 1918. SbEtEdáq, a shamanistic performance of the Coast Salish.

American Anthropologist 20, 249–257.

Haeberlin, Herman. (n.d.) Unpublished Puget Sound Salish texts. Seattle, University of

Washington Libraries.

Hagiwara, Robert E. (1987). Lushootseed copular and Wh–deixis in a Government and Binding

model of Grammar. 22

International Conference on Salish and Neighbouring Languages,

101–108. Victoria, B.C.

564

Hagiwara, Robert E. (1989). Pronominal arguments and the syntax of Lushootseed transitives.

International Conference on Salish and Neighbouring Languages, 61–75. Steilacoom,

Washington.

Hagiwara, Robert E. (1990). Lushootseed (Salish) transitives, pronominal morphology and

licensing of noun phrases. M.A. thesis, University of California, Los Angeles.

Halliday, Michael A. K. (1970) Language structure and language function. In John Lyons (ed.),

New horizons in linguistics, 140–165. Harmondsworth: Penguin

Harrington, John P. (1910). Duwamish fieldnotes. Microfilm, reel No. 015, John Peabody

Harrington Papers, Alaska/Northwest Coast, National Anthropology Archives, Smithsonian

Institution, Washington, D.C.

Hess, Thomas M. (1966). Snohomish chameleon morphology. International Journal of American

Linguistics 30, 350–356.

Hess, Thomas M. (1967a). Snohomish grammatical structure. Seattle: University of Washington.

Hess, Thomas M. (1967b.) The morph /–($)b/ in Snohomish. 2

International Conference on

Salish Languages. Seattle, Washington.

Hess, Thomas M. (1968). Directive phrases, a consideration of one facet of Puget Salish Syntax.

International Conference on Salish Languages. Victoria, B.C.

Hess, Thomas M. (1969). Secondary suffixation in Puget Salish. 4

International Conference on

Salish Languages. Victoria, B.C.

Hess, Thomas M. (1971). Prefix constituents with /xʷ/. In James E. Hoard and Thomas M. Hess

(eds.), Studies in Northwest Indian Languages, 43–69. Sacramento: Sacramento

Anthropological Society.

565

Hess, Thomas M. (1972). Some lexical sets in Puget Salish orientation vocabulary. 7

International Conference on Salish Languages. Bellingham, WA: University of Hawaii

Working Papers in Linguistics.

Hess, Thomas M. (1974). How do you say “You Are Our Father” in Salish? 9

International

Conference on Salish Languages, 53–59. Vancouver, B.C.

Hess, Thomas M. (1976). Dictionary of Puget Salish. Seattle: American Indian Studies,

University of Washington Press.

Hess, Thomas M. (1977). Lushootseed dialects. Anthropological Linguistics 19, 403–419.

Hess, Thomas M. (1979a). A comparison of marine and riverine orientation vocabulary in two

Coast Salish languages. Anthropological Linguistics 21, 363–378.

Hess, Thomas M. (1979c). Central Coast Salish Words for Deer: Their wavelike distribution.

International Journal of American Linguistics 45, 5–16.

Hess, Thomas M. (1982). Traces of “abnormal” speech in Lushootseed. 17

International

Conference on Salishan and Neighboring Languages, 89–97, Portland, Oregon.

Hess, Thomas M. (1986). Borrowed words and prehistory along the Straits of Juan de Fuca.

Working Papers of the Linguistics Circle of the University of Victoria 5 (2), 71–86.

Hess, Thomas M. (1990). Another component of meaning for {-n$xʷ}. Papers for the 25

International Conference on Salish and Neighboring Languages, 173–76. Vancouver:

University of British Columbia.

Hess, Thomas M. (1993). A schema for the presentation of Lushootseed verb stems. In Anthony

Mattina and Timothy R. Montler (eds.), American Indian linguistics and ethnography in

honor of Laurence C. Thompson, 113–126. Missoula, MT: University of Montana

Occasional Papers in Linguistics.

566

Hess, Thom. (1995). Lushootseed reader with introductory grammar, Volume I: Four Stories

from Edward Sam. Missoula: University of Montana Occasional Papers in Linguistics.

Hess, Thom. (1998). Lushootseed reader with intermediate grammar, Volume II: Four Stories

from Martha Lamont. Missoula: University of Montana Occasional Papers in Linguistics.

Hess, Thom. (2006). Lushootseed reader with English Translations, Volume III: Four More

Stories from Martha Lamont. Missoula: University of Montana Occasional Papers in

Linguistics.

Hess, Thomas M., and Dawn Bates. (1998). Semantic role assignment in Lushootseed causatives.

In Dawn Bates (ed.), 33

International Conference on Salish and Neighbouring Languages

221–234. Seattle: University of Washington.

Hess, Thomas M., and Dawn Bates. (2004). Lushootseed applicatives and their ilk. In Donna B.

Gerdts and Lisa Matthewson (eds.), Studies in Salish linguistics in honor of M. Dale

Kinkade., 172-196. Helena, MT: University of Montana.

Hess, Thomas M. and Vi Taqʷš$blu Hilbert. (1976). Lushootseed: An introduction, Books 1 and

2. Seattle: American Indian Studies, University of Washington.

Hilbert, Vi Taqʷš$blu. (1974). On transcribing the Metcalf tapes. 9

International Conference on

Salish Languages, 49–52. Vancouver, B.C.

Hilbert, Vi Taqʷš$blu. (1983). Poking fun in Lushootseed. 18

International Conference on

Salishan and Neighboring Languages, 197–213. Seattle, Washington.

Hilbert, Vi Taqʷš$blu. (1985). Haboo, Native American stories from Puget Sound. Seattle, WA:

University of Washington Press.

567

Hilbert, Vi Taqʷš$blu. (1991). When Chief Seattle (Si–alh) spoke. In Robin Wright (ed.), A time

of gathering, native heritage in Washington State, 259–266. Seattle, University of

Washington Press.

Hilbert, Vi Taqʷš$blu. (1995a). x&əč’usədəʔ ʔ! gʷ!qʷulc’!ʔ, Aunt Susie Sampson Peter: The

wisdom of a Skagit elder. Seattle, WA: Lushootseed Press.

Hilbert, Vi Taqʷš$blu. (1995b). siast!nu, “Gram” Ruth Sehome Shelton: The wisdom of a

Tulalip elder. Seattle, WA: Lushootseed Press.

Hilbert, Vi Taqʷš$blu, and Thomas M. Hess. (1975). A note on $ constructions in Lushootseed.

International Conference on Salish Languages. Ellensburg, Washington.

Hilbert, Vi Taqʷš$blu, and Thomas M. Hess. (1977). Lushootseed. In Barry F. Carlson (ed.),

Northwest Coast texts: Stealing light, 4–32. Chicago: University of Chicago Press.

Hilbert, Vi Taqʷš$blu, and Jay Miller. (2004). That Salish feeling. In Donna B. Gerdts and Lisa

Matthewson (eds.), Studies in Salish linguistics in honor of M. Dale Kinkade, 197-210.

Helena, MT: University of Montana.

Hopper, Paul and Thompson, Sandra A. (1980). Transitivity in grammar and discourse.

Language 56, 251–99.

Hukari, Thomas E. (1976). Person in a Coast Salish Language. International Journal of American

Linguistics 42, 305–318.

Hukari, Thomas E. (1977). A comparison of attributive clause constructions in two Coast Salish

languages. Glossa 11, 48–73.

Jelinek, Eloise and Demers, Richard A. (1983). The agent hierarchy and voice in some Coast

Salish languages. International Journal of American Linguistics 49, 167–85.

568

Kaisse, Ellen. (1980). Formalizing the assignment of vowel height in Lushootseed. 15

International Conference on Salishan and Neighboring Languages 15, 163–173,

Vancouver, B.C.

Kemmer, Suzanne (1993). The middle voice. Amsterdam: John Benjamins.

Kim, Hyong Joong. (1995). Korean and Lushootseed Salish from a functional perspective. PhD.

dissertation, University of Victoria.

Kinkade, M. Dale (1990). Sorting out third persons in Salish discourse. International Journal of

American Linguistics 56, 341–360.

Kirkham, Sandra P. 1992. Reduplication in Lushootseed: A prosodic analysis. M.A. thesis,

University of Victoria.

Kroeber, Paul D. (1999). The Salishan language family: Reconstructing syntax. Lincoln, NB:

University of Nebraska Press.

Lambrecht, Knud (1994). Information structure and sentence form: Topic, focus, and the mental

representation of discourse referents. Cambridge: Cambridge University Press.

Langacker, Ronald W. (1987). Foundations of cognitive grammar. Stanford, Calif.: Stanford

University Press.

Langen, Toby C. S. 1984. Four Upper Skagit Versions of “Starchild.” 19

International

Conference on Salish and Neighbouring Languages, 241–254. Victoria, B.C.

Langen, Toby C. S. 1985. Creativity with Tradition: Susan Sampson Peter at Raven’s Feast. 20

International Conference on Salish and Neighbouring Languages, 197–212. Vancouver,

B.C.

Langen, Toby C. S. 1986. Notes on form in some Northwest Coast Tales. 21

International

Conference on Salish and Neighbouring Languages, 123–134. Seattle, Washington.

569

Langen, Toby C. S. 1988. Mrs. Peter’s Bluejay Story: Narrative Modes. 23

International

Conference on Salish and Neighbouring Languages, 50–77. Eugene, Oregon.

Langen, Toby C. S. 1990. How Long Does “Mythification” Take? Thoughts on a Lushootseed

Story about a Disabled Boy. 25

International Conference on Salish and Neighbouring

Languages, 225–232. Vancouver, B.C.

Langen, Toby C. S. 1992. ‘Translating Form in Classical American Indian Literature.’ In Brian

Swann (ed.), On the Translation of Native American Literatures, 191–207. Washington,

D.C.: Smithsonian Institution

Langen, Toby C. S. 1997. On the Predictability of Martha Lamont’s “Pheasant and Raven.” 32

International Conference on Salish and Neighbouring Languages,280-290, Port Angeles,

Washington. [Also published as pp. 164-174 of Hess 1998.]

Langen, Toby C. S. 1998a. Hermeneutic Functions of Style in Martha Lamont’s “Mink and

Changer.” 33

International Conference on Salish and Neighbouring Languages, 335–346.

Seattle, Washington.

Langen, Toby C. S. 1998b. On the Predictability of Martha Lamont’s “Pheasant and Raven.” In

Thomas M. Hess, Lushootseed reader, Volume II, 164–174. Missoula: University of

Montana Occasional Papers in Linguistics.

Langen, Toby C. S. 1999. Parody and Interrogation in Martha Lamont’s Two Tellings of “Crow

is Sick.” 34

International Conference on Salish and Neighbouring Languages, 253–264.

Kamloops, B.C.

Lonsdale, Deryle. (2001). A two–level implementation for Lushootseed morphology. In Leora

Bar-El, Linda T. Watt, and Ian Wilson (eds.), Papers for ICSNL XXVI: The Thirty-sixth

570

International Conference on Salish and Neighbouring Languages, 203–214. Vancouver:

University of British Columbia Working Papers in Linguistics.

Lonsdale, Deryle. (2002). A Categorial Grammar fragment for Lushootseed. In Carrie Gillon, N.

Sawai, and Rachel Wojdak (eds.), Papers for ICSNL XXXVII: The Thirty Seventh

International Conference on Salish and Neighboring Languages, 215–232. Vancouver:

University of British Columbia Working Papers in Linguistics.

Lonsdale, Deryle. (2003). Doing Lushootseed morphology by analogy. In J. C. Brown and

Michele Kalmar (eds.), Papers for ICSNL XXVIII: The Thirty-eighth International

Conference on Salishan and Neighboring Languages, 175–184. Vancouver: University of

British Columbia Working Papers in Linguistics.

MacLaury, Robert E., and Brent D. Galloway. (1988). Color categories and color qualifiers in

Halkomelem, Samish, Lushootseed, Nooksack, and Yakima. 23

International Conference

on Salishan and Neighboring Languages, 166–199. Eugene, Oregon.

Matthewson, Lisa (1996). Determiner systems and quantificational strategies: Evidence from

Salish. PhD dissertation, University of British Columbia.

Mel’čuk, Igor A. (1993-2000). Cours de morphologie générale. Montreal: Presses de l'Université

de Montréal.

Mel’čuk, Igor A. (2001). Communicative organization in natural language . Amsterdam: John

Benjamins.

Mel’čuk, Igor A. (2006). Aspects of the theory of morphology. Berlin: Mouton de Gruyter.

Montler, Timothy R. (1996). Some Klallam paradigms. International Conference on Salishan and

Neighbouring Languages 257–263. Vancouver, B.C.

571

Moses, Marya, and Toby C. S. Langen. (2001). Reading Martha Lamont’s Crow story today. In,

Larry Evers and Barre Toelken (eds.), Native American oral traditions, collaboration and

interpretation. Logan: Utah State University Press.

Nater, H. F. (1984). The Bella Coola language. Ottawa: National Museums of Canada.

Newman, Stanley (1969). Bella Coola Paradigms. International Journal of American Linguistics

35, 299-306.

Newman, Stanley (1977). The Salish independent pronoun system. International Journal of

American Linguistics 43, 302-314.

Newman, Stanley (1979). A history of the Salish possessive and subject forms. International

Journal of American Linguistics 45, 207-223.

Newman, Stanley (1980). Functional changes in the Salish pronominal system. International

Journal of American Linguistics 46, 155-167.

Park, Miae. (2000). Surface opacity and phonological issues in Klamath and Lushootseed. Ph.D.

dissertation, University of Wisconsin – Madison.

Ransom, Jay E. (1945). Notes on Duwamish phonology and morphology. International Journal

of American Linguistics 11, 204–210.

Sapir, Edward (2004) Language: An introduction to the study of speech. New York: Dover

Smith, Marian W. (n.d.) Vocabularies in Nootsack and other Coast Salishan Languages. MS S8,

Boas Collection, American Philosophical Society Library, Philadelphia, Pennsylvania.

Snyder, Warren A. (1968a). Southern Puget Sound Salish: Phonology and Morphology.

Sacramento Anthropological Society Paper 8.

Snyder, Warren A. (1968b). Southern Puget Sound Salish: Texts, place names and dictionary.

Sacramento Anthropological Society Paper 9.

572

Stassen, Leon. (1984). The comparative compared. Journal of Semantics 3, 143–182.

Thompson, Laurence C. and Thompson, M. Terry (1992). The Thompson language. Missoula,

MT: University of Montana Occasional Papers in Linguistics.

Thompson, Nile R. (1981). A contribution to Puget Salish lexical studies. In Anthony Mattina

and Timothy R. Montler (eds.), 14

International Conference on Salish and Neighboring

Languages, 80–83. Missoula, MT.

Thompson, Nile R., and James L. Armagost. (1996). The Agent–Patient restriction in Puget

Sound Salish. 31

International Conference on Salishan and Neighboring Languages.

Vancouver, B.C.

Tweddell, Colin Ellidge. (1950). The Snoqualmie-Duwamish dialects of Puget Sound Salish: An

outline of phonemics and morphology. University of Washington Publications in

Anthropology 12. Seattle, University of Washington Press.

Urbanczyk, Suzanne C. (1996a). Patterns of Reduplication in Lushootseed. PhD dissertation,

University of Massachusetts, Amherst.

Urbanczyk, Suzanne C. (1996b). Preliminary remarks on Lushootseed syncope. 31

International Conference on Salishan and Neighboring Languages, 321–326. Vancouver,

B.C.

van Eijk, Jan P. (1998). The morphological status of partial reduplication: Evidence from

Lushootseed and Lillooet. In Mark Janse (ed.), Productivity and creativity: Studies in

General and Descriptive Linguistics in Honor of E. M. Uhlenbeck, 433–448. Berlin: Mouton

de Gruyter.

van Eijk, Jan P. (n.d.). An annotated bibliography of Salish Linguistics. Regina, SK: First

Nations University.

573

Winter, Anne. (1996). Transitivity and intransitivity in Lushootseed. M.A. thesis, Arizona Sate

University.